{ "meta": { "source": "atlas_v13_current.html", "snapshot": "2026-10-18", "site_version": "13.2", "count": 111, "note": "Materialized view of the current audited graph; complete node fields retained. Source inspection and publication status are recorded per item. The builder performs no new empirical validation." }, "records": [ { "id": "B001", "label": "Introduzioni alle parti — Non-Deterministic Software Engineering", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Productivity", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C01", "C:C11" ], "printed": false, "summary": "Le introduzioni alle parti costruiscono il percorso dalle trasformazioni professionali all’adozione organizzativa.", "limitation": "La funzione è orientativa: la copertura del volume non va conteggiata come evidenza aggiuntiva rispetto ai capitoli figli.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Introduzioni alle parti — Non-Deterministic Software Engineering", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/main.tex", "source_sha256": "bcfc9816cbf334fa7890857dcc1bd80c4b70a1ee0daad84c0ef957bb818271ed", "line_count": 241, "word_count_approx": 1126, "card_type": "structural_synthesis", "card_path": "cards/V01/B001.md", "in_short": "Le introduzioni alle parti costruiscono il percorso dalle trasformazioni professionali all’adozione organizzativa.", "summary": "Il main contiene prosa autonoma di raccordo fra fondamenti, implementazione, organizzazione, catalogo e appendici. Va schedato senza inglobare nuovamente il testo dei file inclusi.", "limitation": "La funzione è orientativa: la copertura del volume non va conteggiata come evidenza aggiuntiva rispetto ai capitoli figli.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Productivity\", \"T:Skills & learning\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Human-agent collaboration\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C11\"]", "evidence_anchors_json": "[{\"title\": \"Introduzioni alle parti — Non-Deterministic Software Engineering\", \"line\": 1}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Human-agent collaboration", "K:Evaluation methodology and baselines" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/main.tex" } }, { "id": "B002", "label": "Fonti, attribuzioni e avvertenze di lettura", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do" ], "actual": [], "mapped": [ "C:C01" ], "printed": false, "summary": "Il frontespizio dichiara fonti, attribuzioni e limiti del volume.", "limitation": "Attribuzioni generali e disclaimer non dimostrano l’identità di ogni citazione né l’accuratezza dei dati riportati.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Fonti, attribuzioni e avvertenze di lettura", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/frontmatter.tex", "source_sha256": "5a3ff2b34f5ee94776cf200a08fb274ccd607a34b6cf07d0945dd3f62cf3e1a0", "line_count": 170, "word_count_approx": 379, "card_type": "reading_guide", "card_path": "cards/V01/B002.md", "in_short": "Il frontespizio dichiara fonti, attribuzioni e limiti del volume.", "summary": "Le note iniziali chiariscono l’uso di interviste practitioner, report industriali ed esperienza dell’autore; permettono di qualificare l’origine delle affermazioni successive.", "limitation": "Attribuzioni generali e disclaimer non dimostrano l’identità di ogni citazione né l’accuratezza dei dati riportati.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Human-agent collaboration\"]", "editorial_chapters_json": "[\"C:C01\"]", "evidence_anchors_json": "[{\"title\": \"Fonti, attribuzioni e avvertenze di lettura\", \"line\": 1}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Human-agent collaboration" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/frontmatter.tex" } }, { "id": "B003", "label": "Summary", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Productivity", "T:Economics", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C01", "C:C02", "C:C11" ], "printed": false, "summary": "Il sommario introduce il problema della distanza fra feature generata e cambiamento pronto per il rilascio.", "limitation": "Il caso iniziale e i tempi di lettura sono dispositivi didattici, non risultati sperimentali.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Summary", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/summary_v2.tex", "source_sha256": "d249bc8ae2f50a1eb6fee7a928b83ea9d7593c71a7c2d4af3f681ab1e63ca577", "line_count": 326, "word_count_approx": 3214, "card_type": "synthesis", "card_path": "cards/V01/B003.md", "in_short": "Il sommario introduce il problema della distanza fra feature generata e cambiamento pronto per il rilascio.", "summary": "La scena iniziale e i percorsi per ruolo spiegano come usare il libro. La scheda è un nodo di orientamento che collega trasformazione del lavoro, misurazione e responsabilità.", "limitation": "Il caso iniziale e i tempi di lettura sono dispositivi didattici, non risultati sperimentali.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Productivity\", \"T:Economics\", \"T:Skills & learning\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Human-agent collaboration\", \"K:Output versus shipped value\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C02\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Opening Scenario\", \"line\": 13, \"offset\": 466, \"end\": 496}, {\"level\": \"section\", \"title\": \"Why This Book Exists\", \"line\": 84, \"offset\": 7748, \"end\": 7778}, {\"level\": \"section\", \"title\": \"What You'll Learn\", \"line\": 122, \"offset\": 11432, \"end\": 11459}, {\"level\": \"section\", \"title\": \"Reading Paths\", \"line\": 227, \"offset\": 21738, \"end\": 21761}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Human-agent collaboration", "K:Output versus shipped value", "K:Evaluation methodology and baselines" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/summary_v2.tex" } }, { "id": "B004", "label": "The Fundamental Shift", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C01", "C:C06" ], "printed": false, "summary": "Il passaggio al non determinismo richiede pratiche di controllo diverse dalla sola produzione del codice.", "limitation": "L’analogia con precedenti salti di astrazione è interpretativa; non dimostra che tutti gli strumenti abbiano lo stesso profilo di affidabilità.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Fundamental Shift", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter01_v4.tex", "source_sha256": "ad460f2b88bac8835e3bc1dab99bd1aefcac228172ab226fe9eb063dd7396a81", "line_count": 257, "word_count_approx": 2767, "card_type": "chapter", "card_path": "cards/V01/B004.md", "in_short": "Il passaggio al non determinismo richiede pratiche di controllo diverse dalla sola produzione del codice.", "summary": "Il capitolo propone una lettura storica del cambiamento e ne ricava responsabilità di verifica, modelli mentali e attenzione alla sicurezza.", "limitation": "L’analogia con precedenti salti di astrazione è interpretativa; non dimostra che tutti gli strumenti abbiano lo stesso profilo di affidabilità.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Code quality & security\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Human-agent collaboration\", \"K:Prompt injection and malicious tools\", \"K:Vulnerability and secure code\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C06\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"A Shift as Significant as Assembly to High-Level Languages\", \"line\": 13, \"offset\": 706, \"end\": 774}, {\"level\": \"section\", \"title\": \"Implications for Engineering Practice\", \"line\": 80, \"offset\": 8516, \"end\": 8563}, {\"level\": \"section\", \"title\": \"The Security Imperative\", \"line\": 116, \"offset\": 11812, \"end\": 11845}, {\"level\": \"section\", \"title\": \"Why This Historical Context Matters\", \"line\": 214, \"offset\": 18860, \"end\": 18905}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Human-agent collaboration", "K:Prompt injection and malicious tools", "K:Vulnerability and secure code" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter01_v4.tex" } }, { "id": "B005", "label": "Vibe Coding vs. AI-Assisted Engineering", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Verification & review", "T:Skills & learning", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C01", "C:C04", "C:C10" ], "printed": false, "summary": "Vibe coding e ingegneria assistita differiscono per comprensione, controllo e responsabilità.", "limitation": "Il 70% è un’euristica practitioner: non deve diventare un tasso universale di completamento o correttezza.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Vibe Coding vs. AI-Assisted Engineering", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter02_v3.tex", "source_sha256": "abb747928eee306738180a372fd5226ccba87eac886e1c3ea22f88ee6e17acb7", "line_count": 365, "word_count_approx": 3298, "card_type": "chapter", "card_path": "cards/V01/B005.md", "in_short": "Vibe coding e ingegneria assistita differiscono per comprensione, controllo e responsabilità.", "summary": "Il testo delimita l’uso esplorativo dei prototipi, discute il problema del 70% e introduce specifiche e verifiche come alternativa alla mera prosecuzione della conversazione.", "limitation": "Il 70% è un’euristica practitioner: non deve diventare un tasso universale di completamento o correttezza.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Verification & review\", \"T:Skills & learning\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Human-agent collaboration\", \"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Cognitive offloading and skill debt\", \"K:System understanding and rationale\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C04\", \"C:C10\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Two Fundamentally Different Approaches\", \"line\": 21, \"offset\": 1676, \"end\": 1724}, {\"level\": \"section\", \"title\": \"When Vibes Meet Reality\", \"line\": 79, \"offset\": 7228, \"end\": 7261}, {\"level\": \"section\", \"title\": \"The Spec-Driven Alternative\", \"line\": 209, \"offset\": 16189, \"end\": 16226}, {\"level\": \"section\", \"title\": \"Organizational Implications\", \"line\": 308, \"offset\": 23405, \"end\": 23442}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Human-agent collaboration", "K:Specifications and governance", "K:Independent verification and oracles", "K:Cognitive offloading and skill debt", "K:System understanding and rationale", "K:Ownership and knowledge transfer" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter02_v3.tex" } }, { "id": "B006", "label": "The Experience Divide", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Skills & learning", "T:System knowledge & memory", "T:Jobs & roles" ], "actual": [], "mapped": [ "C:C10", "C:C12" ], "printed": false, "summary": "L’esperienza cambia il modo in cui si delega all’AI, ma va preservato il ciclo di apprendimento.", "limitation": "Le classi di esperienza non sono gruppi sperimentali; le prescrizioni formative richiedono misure di comprensione, retention e trasferimento.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Experience Divide", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter03_v3.tex", "source_sha256": "2c277baece8dabb6c495ca69c1c29803ec769bf2f0a9e0054bf3f73794a839b8", "line_count": 392, "word_count_approx": 3477, "card_type": "chapter", "card_path": "cards/V01/B006.md", "in_short": "L’esperienza cambia il modo in cui si delega all’AI, ma va preservato il ciclo di apprendimento.", "summary": "Il capitolo analizza falsa sicurezza, trio programming e trasformazione delle competenze per junior, intermedi e senior. Offre un ponte fra produttività individuale e formazione del team.", "limitation": "Le classi di esperienza non sono gruppi sperimentali; le prescrizioni formative richiedono misure di comprensione, retention e trasferimento.", "primary_theme": "T:Skills & learning", "streams_json": "[\"T:Skills & learning\", \"T:System knowledge & memory\", \"T:Jobs & roles\"]", "topics_json": "[\"K:Skills formation and teaching\", \"K:Cognitive offloading and skill debt\", \"K:Employment and early-career roles\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C10\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Why Experience Matters More Than Ever\", \"line\": 27, \"offset\": 2901, \"end\": 2948}, {\"level\": \"section\", \"title\": \"Preserving the Learning Loop\", \"line\": 95, \"offset\": 8406, \"end\": 8444}, {\"level\": \"section\", \"title\": \"The Skill Transformation\", \"line\": 221, \"offset\": 18000, \"end\": 18034}, {\"level\": \"section\", \"title\": \"Control, Confidence, and Understanding\", \"line\": 356, \"offset\": 25951, \"end\": 25999}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Skills formation and teaching", "K:Cognitive offloading and skill debt", "K:Employment and early-career roles", "K:Ownership and knowledge transfer" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter03_v3.tex" } }, { "id": "B007", "label": "New Workflows for the AI Era", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Verification & review" ], "actual": [], "mapped": [ "C:C03", "C:C04", "C:C07", "C:C11" ], "printed": false, "summary": "Workflow piccoli e verificabili trasformano l’AI da generatore a componente del processo ingegneristico.", "limitation": "Le descrizioni dei workflow non misurano automaticamente un vantaggio di velocità o qualità; mantenere le dipendenze fra i controlli.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "New Workflows for the AI Era", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter04_v5.tex", "source_sha256": "763822b37c388a0fe474f8e30be03eb08bc48ecdae910167bb8116e5c2b6b2f3", "line_count": 987, "word_count_approx": 7757, "card_type": "chapter", "card_path": "cards/V01/B007.md", "in_short": "Workflow piccoli e verificabili trasformano l’AI da generatore a componente del processo ingegneristico.", "summary": "Pianificazione, TDD, runtime feedback, contesto, review e orchestrazione enterprise sono raccolti in un percorso operativo. Il file è il principale punto di origine di varie elaborazioni dei volumi successivi.", "limitation": "Le descrizioni dei workflow non misurano automaticamente un vantaggio di velocità o qualità; mantenere le dipendenze fra i controlli.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Verification & review\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C04\", \"C:C07\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Workflow: Plan Before You Prompt\", \"line\": 17, \"offset\": 1293, \"end\": 1335}, {\"level\": \"section\", \"title\": \"Workflow: Keeping the Runtime in the Loop\", \"line\": 215, \"offset\": 17257, \"end\": 17308}, {\"level\": \"section\", \"title\": \"Workflow: Understanding Legacy Systems\", \"line\": 549, \"offset\": 42402, \"end\": 42450}, {\"level\": \"section\", \"title\": \"The Orchestration Mindset\", \"line\": 920, \"offset\": 68678, \"end\": 68713}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Specifications and governance", "K:Independent verification and oracles", "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter04_v5.tex" } }, { "id": "B008", "label": "The Quality Crisis (And How to Prevent It)", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Verification & review", "T:Code quality & security", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C05", "C:C08" ], "printed": false, "summary": "La qualità diventa critica quando il volume generato supera la capacità di revisione.", "limitation": "Distinguere difetti dimostrati, segnalazioni statiche, percezioni dei revisori e spiegazioni causali.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Quality Crisis (And How to Prevent It)", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter05_v4.tex", "source_sha256": "31cb9350dc88aef0f7ac2f8ef20be581d212c82b99c42be925d514dd90782ad2", "line_count": 310, "word_count_approx": 4277, "card_type": "chapter", "card_path": "cards/V01/B008.md", "in_short": "La qualità diventa critica quando il volume generato supera la capacità di revisione.", "summary": "Il capitolo combina debito tecnico, fiducia, colli di bottiglia e quality gate; invita a rendere verificabile l’accettazione e a mantenere una cultura di pensiero critico.", "limitation": "Distinguere difetti dimostrati, segnalazioni statiche, percezioni dei revisori e spiegazioni causali.", "primary_theme": "T:Verification & review", "streams_json": "[\"T:Verification & review\", \"T:Code quality & security\", \"T:Engineering structure\"]", "topics_json": "[\"K:Review capacity and maintainers\", \"K:AI as reviewer\", \"K:Automation bias and trust\", \"K:Architecture and modularity\", \"K:Technical debt and complexity\"]", "editorial_chapters_json": "[\"C:C05\", \"C:C08\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Compounding Effect of Volume\", \"line\": 20, \"offset\": 1186, \"end\": 1228}, {\"level\": \"section\", \"title\": \"The Trust Problem\", \"line\": 80, \"offset\": 8826, \"end\": 8853}, {\"level\": \"section\", \"title\": \"The Critical Thinking Imperative\", \"line\": 213, \"offset\": 22839, \"end\": 22881}, {\"level\": \"section\", \"title\": \"The Path Forward\", \"line\": 271, \"offset\": 29784, \"end\": 29810}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Review capacity and maintainers", "K:AI as reviewer", "K:Automation bias and trust", "K:Architecture and modularity", "K:Technical debt and complexity" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter05_v4.tex" } }, { "id": "B009", "label": "Measuring What Matters---The Data Behind AI's Real Impact", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Productivity", "T:Economics", "T:Verification & review" ], "actual": [], "mapped": [ "C:C02", "C:C12" ], "printed": false, "summary": "La misura dell’AI deve includere il lavoro rilasciato, la qualità e l’esperienza dello sviluppatore.", "limitation": "Il riferimento a 180+ organizzazioni non rende omogenee metriche, campioni o disegni; preservare denominatori e possibili interessi delle fonti.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Measuring What Matters---The Data Behind AI's Real Impact", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter06_v4.tex", "source_sha256": "3e5d40ca87c35ee64cb7f0dd7b55a9d07f12e00738955651fa8a57875bde4bc9", "line_count": 367, "word_count_approx": 4794, "card_type": "chapter", "card_path": "cards/V01/B009.md", "in_short": "La misura dell’AI deve includere il lavoro rilasciato, la qualità e l’esperienza dello sviluppatore.", "summary": "Il testo confronta narrazioni di produttività, dati industriali, soddisfazione, dimensione delle modifiche e vantaggi nella riduzione del lavoro ripetitivo.", "limitation": "Il riferimento a 180+ organizzazioni non rende omogenee metriche, campioni o disegni; preservare denominatori e possibili interessi delle fonti.", "primary_theme": "T:Productivity", "streams_json": "[\"T:Productivity\", \"T:Economics\", \"T:Verification & review\"]", "topics_json": "[\"K:Output versus shipped value\", \"K:Evaluation methodology and baselines\", \"K:Controlled productivity evidence\"]", "editorial_chapters_json": "[\"C:C02\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Understanding the Hype Problem\", \"line\": 15, \"offset\": 1042, \"end\": 1082}, {\"level\": \"section\", \"title\": \"The Surprising Hierarchy of Time Savings\", \"line\": 164, \"offset\": 16180, \"end\": 16230}, {\"level\": \"section\", \"title\": \"Hidden Benefits Beyond Time Savings\", \"line\": 250, \"offset\": 25126, \"end\": 25171}, {\"level\": \"section\", \"title\": \"Practical Guidance for Engineering Leaders\", \"line\": 308, \"offset\": 32822, \"end\": 32874}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Output versus shipped value", "K:Evaluation methodology and baselines", "K:Controlled productivity evidence" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter06_v4.tex" } }, { "id": "B010", "label": "The Adaptation Imperative---What Companies Must Do Now", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Productivity", "T:Skills & learning", "T:Verification & review" ], "actual": [], "mapped": [ "C:C11", "C:C12" ], "printed": false, "summary": "L’adozione richiede una trasformazione di processi e competenze, non soltanto licenze.", "limitation": "La roadmap è una proposta operativa; tempi e ritorni devono essere calibrati sul contesto anziché assunti come dimostrati.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Adaptation Imperative---What Companies Must Do Now", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter08_v4.tex", "source_sha256": "5682d0823509380c14528b7d07abb1cccd56e5ecd396072b04697db07c93cc83", "line_count": 433, "word_count_approx": 4765, "card_type": "chapter", "card_path": "cards/V01/B010.md", "in_short": "L’adozione richiede una trasformazione di processi e competenze, non soltanto licenze.", "summary": "Otto principi, una roadmap di sei mesi e adattamenti per dimensione aziendale traducono il cambiamento tecnico in responsabilità e decisioni organizzative.", "limitation": "La roadmap è una proposta operativa; tempi e ritorni devono essere calibrati sul contesto anziché assunti come dimostrati.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Productivity\", \"T:Skills & learning\", \"T:Verification & review\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C11\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Contrasting Approaches: Speed Versus Rigor\", \"line\": 15, \"offset\": 1182, \"end\": 1234}, {\"level\": \"section\", \"title\": \"Measuring AI's Real Impact\", \"line\": 180, \"offset\": 14624, \"end\": 14660}, {\"level\": \"section\", \"title\": \"Your 6-Month Roadmap\", \"line\": 190, \"offset\": 15973, \"end\": 16003}, {\"level\": \"section\", \"title\": \"For Individual Contributors\", \"line\": 402, \"offset\": 34084, \"end\": 34121}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Evaluation methodology and baselines", "K:Specifications and governance", "K:Ownership and knowledge transfer" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter08_v4.tex" } }, { "id": "B011", "label": "Looking Forward---The Future of Software Engineering", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Jobs & roles", "T:Skills & learning", "T:What agents can do", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C01", "C:C11", "C:C12" ], "printed": false, "summary": "Le traiettorie future sono scenari da usare per prepararsi, non previsioni certe.", "limitation": "Conservare l’ancoraggio temporale dell’edizione: prospettive a 12–18 mesi o a più anni non sono fatti accertati oggi.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Looking Forward---The Future of Software Engineering", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter09_v4.tex", "source_sha256": "3cc847ac00e2eb21da6b48122b67386387ba80b81c12ba0549e1a18b91959e0d", "line_count": 373, "word_count_approx": 4375, "card_type": "chapter", "card_path": "cards/V01/B011.md", "in_short": "Le traiettorie future sono scenari da usare per prepararsi, non previsioni certe.", "summary": "Il capitolo articola orizzonti temporali e implicazioni di carriera, distinguendo possibilità favorevoli e rischi per la professione.", "limitation": "Conservare l’ancoraggio temporale dell’edizione: prospettive a 12–18 mesi o a più anni non sono fatti accertati oggi.", "primary_theme": "T:Jobs & roles", "streams_json": "[\"T:Jobs & roles\", \"T:Skills & learning\", \"T:What agents can do\", \"T:Engineering structure\"]", "topics_json": "[\"K:Employment and early-career roles\", \"K:Ownership and knowledge transfer\", \"K:Agent failure and escalation\", \"K:Long-running agent execution\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C11\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Near Future: 12--18 Months\", \"line\": 15, \"offset\": 1232, \"end\": 1272}, {\"level\": \"section\", \"title\": \"The Long Term: 5+ Years\", \"line\": 227, \"offset\": 17261, \"end\": 17294}, {\"level\": \"section\", \"title\": \"The Fork in the Road\", \"line\": 295, \"offset\": 26884, \"end\": 26914}, {\"level\": \"section\", \"title\": \"Your Next Steps\", \"line\": 321, \"offset\": 30594, \"end\": 30619}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Employment and early-career roles", "K:Ownership and knowledge transfer", "K:Agent failure and escalation", "K:Long-running agent execution" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter09_v4.tex" } }, { "id": "B012", "label": "The Pattern Catalog", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do" ], "actual": [], "mapped": [ "C:C01", "C:C04", "C:C11" ], "printed": false, "summary": "Il catalogo organizza pattern riusabili e dipendenze fra pratiche.", "limitation": "Non duplicare il contenuto dei figli; l’espressione pattern non implica validazione comparativa della sua efficacia.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Pattern Catalog", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter07_v1.tex", "source_sha256": "884c52d5f39189d0a3b60e77f42ddec5d371362d59f7fb84a8a4d4678cfe9f52", "line_count": 193, "word_count_approx": 713, "card_type": "chapter", "card_path": "cards/V01/B012.md", "in_short": "Il catalogo organizza pattern riusabili e dipendenze fra pratiche.", "summary": "Il file introduce importanza, selezione e uso del catalogo; le descrizioni complete dei pattern si trovano nei sei file inclusi e restano schede autonome.", "limitation": "Non duplicare il contenuto dei figli; l’espressione pattern non implica validazione comparativa della sua efficacia.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Specifications and governance\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Understanding Pattern Importance\", \"line\": 13, \"offset\": 611, \"end\": 654}, {\"level\": \"section\", \"title\": \"How to Use This Catalog\", \"line\": 34, \"offset\": 1808, \"end\": 1842}, {\"level\": \"section\", \"title\": \"Complete Pattern List\", \"line\": 49, \"offset\": 2358, \"end\": 2390}, {\"level\": \"section\", \"title\": \"Quick Start Guide\", \"line\": 102, \"offset\": 4604, \"end\": 4632}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Specifications and governance" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter07_v1.tex" } }, { "id": "B013", "label": "PART I: Core Practices", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:Code quality & security", "T:Skills & learning", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C04", "C:C07", "C:C10" ], "printed": false, "summary": "Le pratiche di base uniscono specifica, test, piccoli passi, contesto e comprensione.", "limitation": "Saper spiegare un risultato non è una prova sufficiente di correttezza; la verifica comportamentale deve restare separata.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "PART I: Core Practices", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter07/chapter07_part1.tex", "source_sha256": "d18e91e3e4da41faaeb2ab3780e8680a9360f3b04dc48cfdf64dd2c706d89219", "line_count": 1462, "word_count_approx": 5604, "card_type": "section", "card_path": "cards/V01/B013.md", "in_short": "Le pratiche di base uniscono specifica, test, piccoli passi, contesto e comprensione.", "summary": "I primi otto pattern comprendono spec-first, TDD, esplorazione, chunk verificabili, context engineering, explainability test, runtime feedback e tutela dell’apprendimento.", "limitation": "Saper spiegare un risultato non è una prova sufficiente di correttezza; la verifica comportamentale deve restare separata.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:Code quality & security\", \"T:Skills & learning\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Evaluation methodology and baselines\", \"K:Skills formation and teaching\", \"K:Cognitive offloading and skill debt\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C07\", \"C:C10\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"PART I: Core Practices\", \"line\": 3, \"offset\": 2, \"end\": 35}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:Evaluation methodology and baselines", "K:Skills formation and teaching", "K:Cognitive offloading and skill debt" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter07/chapter07_part1.tex" } }, { "id": "B014", "label": "PART II: Quality Assurance", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Code quality & security", "T:What agents can do", "T:Engineering structure", "T:Verification & review", "T:System knowledge & memory", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C05", "C:C06", "C:C08", "C:C09" ], "printed": false, "summary": "I pattern di assurance collegano sicurezza, revisione graduata e comprensione del legacy.", "limitation": "Una checklist o una review multi-pass non costituisce da sola un controllo indipendente; contano input, oracoli e copertura.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "PART II: Quality Assurance", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter07/chapter07_part2.tex", "source_sha256": "ddaab92e52b76bea22a86d4c6fad3960863640c1ad8eca08d599e9aeb4ff6843", "line_count": 1077, "word_count_approx": 3505, "card_type": "section", "card_path": "cards/V01/B014.md", "in_short": "I pattern di assurance collegano sicurezza, revisione graduata e comprensione del legacy.", "summary": "Security-first, review gauntlet e legacy navigation organizzano controlli e responsabilità prima dell’approvazione dei cambiamenti.", "limitation": "Una checklist o una review multi-pass non costituisce da sola un controllo indipendente; contano input, oracoli e copertura.", "primary_theme": "T:Code quality & security", "streams_json": "[\"T:Code quality & security\", \"T:What agents can do\", \"T:Engineering structure\", \"T:Verification & review\", \"T:System knowledge & memory\", \"T:Skills & learning\"]", "topics_json": "[\"K:Prompt injection and malicious tools\", \"K:Vulnerability and secure code\", \"K:Review capacity and maintainers\", \"K:AI as reviewer\", \"K:Automation bias and trust\", \"K:System understanding and rationale\", \"K:Onboarding and developer fluency\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C05\", \"C:C06\", \"C:C08\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"PART II: Quality Assurance\", \"line\": 3, \"offset\": 2, \"end\": 39}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Prompt injection and malicious tools", "K:Vulnerability and secure code", "K:Review capacity and maintainers", "K:AI as reviewer", "K:Automation bias and trust", "K:System understanding and rationale", "K:Onboarding and developer fluency", "K:Ownership and knowledge transfer" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter07/chapter07_part2.tex" } }, { "id": "B015", "label": "PART III: Team Workflows", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Skills & learning", "T:System knowledge & memory", "T:What agents can do", "T:Engineering structure", "T:Verification & review" ], "actual": [], "mapped": [ "C:C03", "C:C10", "C:C11" ], "printed": false, "summary": "Collaborazione e parallelismo funzionano solo con capacità di apprendimento e revisione adeguate.", "limitation": "Il numero ottimale di agenti dipende da dipendenze, carico di integrazione e attenzione umana; non è una costante universale.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "PART III: Team Workflows", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter07/chapter07_part3.tex", "source_sha256": "76e3939a42d722b1d5a0de3290dd0625337e5a3ffc5fe0e27374334e5ecbc7eb", "line_count": 953, "word_count_approx": 3500, "card_type": "section", "card_path": "cards/V01/B015.md", "in_short": "Collaborazione e parallelismo funzionano solo con capacità di apprendimento e revisione adeguate.", "summary": "Il file tratta migrazione come occasione di apprendimento, trio programming e coordinamento parallelo; discute il rischio della moltiplicazione incontrollata degli agenti.", "limitation": "Il numero ottimale di agenti dipende da dipendenze, carico di integrazione e attenzione umana; non è una costante universale.", "primary_theme": "T:Skills & learning", "streams_json": "[\"T:Skills & learning\", \"T:System knowledge & memory\", \"T:What agents can do\", \"T:Engineering structure\", \"T:Verification & review\"]", "topics_json": "[\"K:Skills formation and teaching\", \"K:Cognitive offloading and skill debt\", \"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C10\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"PART III: Team Workflows\", \"line\": 3, \"offset\": 2, \"end\": 37}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Skills formation and teaching", "K:Cognitive offloading and skill debt", "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter07/chapter07_part3.tex" } }, { "id": "B016", "label": "PART IV: Organizational Foundation", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Productivity", "T:Skills & learning", "T:Verification & review", "T:Economics" ], "actual": [], "mapped": [ "C:C02", "C:C11", "C:C12" ], "printed": false, "summary": "Le fondamenta organizzative rendono espliciti abilitazione, policy, misura e costi.", "limitation": "Esempi aziendali e framework economici non sono repliche indipendenti né garanzie di ROI trasferibile.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "PART IV: Organizational Foundation", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter07/chapter07_part4.tex", "source_sha256": "9d668054284384874a1722dd095b05a02c83d3e3b643c76dc189fa31da25ff1a", "line_count": 1203, "word_count_approx": 3570, "card_type": "section", "card_path": "cards/V01/B016.md", "in_short": "Le fondamenta organizzative rendono espliciti abilitazione, policy, misura e costi.", "summary": "I pattern 15–18 coprono programma di formazione, acceptable use policy, ciclo di misura-apprendimento e gestione della spesa.", "limitation": "Esempi aziendali e framework economici non sono repliche indipendenti né garanzie di ROI trasferibile.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Productivity\", \"T:Skills & learning\", \"T:Verification & review\", \"T:Economics\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\", \"K:Economic costs and return\", \"K:Output versus shipped value\"]", "editorial_chapters_json": "[\"C:C02\", \"C:C11\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"PART IV: Organizational Foundation\", \"line\": 3, \"offset\": 2, \"end\": 47}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Evaluation methodology and baselines", "K:Specifications and governance", "K:Ownership and knowledge transfer", "K:Economic costs and return", "K:Output versus shipped value" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter07/chapter07_part4.tex" } }, { "id": "B017", "label": "Pattern Recap: Quick Reference Guide", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Productivity", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C04", "C:C11" ], "printed": false, "summary": "Il riepilogo aiuta a comporre i pattern secondo situazione e prerequisiti.", "limitation": "È un indice decisionale del catalogo, non un nuovo insieme di risultati; i rimandi ai pattern originari restano espliciti.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Pattern Recap: Quick Reference Guide", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter07/chapter07_part5.tex", "source_sha256": "025bb5b0f700f8a5a904a1c1c0924bf0d4c35f6599dc7df1e5383f5bc1ca39aa", "line_count": 288, "word_count_approx": 1353, "card_type": "section", "card_path": "cards/V01/B017.md", "in_short": "Il riepilogo aiuta a comporre i pattern secondo situazione e prerequisiti.", "summary": "La scheda collega selezione situazionale, mappa delle dipendenze, compatibilità e adozione graduale dei pattern già descritti.", "limitation": "È un indice decisionale del catalogo, non un nuovo insieme di risultati; i rimandi ai pattern originari restano espliciti.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Productivity\", \"T:Skills & learning\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Specifications and governance\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Pattern Recap: Quick Reference Guide\", \"line\": 9, \"offset\": 18, \"end\": 65}, {\"level\": \"section\", \"title\": \"Closing Remarks: Using This Catalog Effectively\", \"line\": 145, \"offset\": 5982, \"end\": 6040}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Specifications and governance", "K:Evaluation methodology and baselines" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter07/chapter07_part5.tex" } }, { "id": "B018", "label": "Antipatterns: What Not to Do", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Verification & review", "T:Code quality & security", "T:Skills & learning", "T:System knowledge & memory", "T:Productivity" ], "actual": [], "mapped": [ "C:C02", "C:C08", "C:C10" ], "printed": false, "summary": "Gli antipattern mostrano come velocità apparente e delega cieca erodano affidabilità e apprendimento.", "limitation": "Il catalogo identifica rischi plausibili; non stima frequenza, severità o effetto causale degli antipattern.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Antipatterns: What Not to Do", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter07/chapter07_antipatterns.tex", "source_sha256": "8cc39f88bf8b3b98d17205a97205e44cc26da2a0ecb460ef0d488908c10043a4", "line_count": 298, "word_count_approx": 1278, "card_type": "section", "card_path": "cards/V01/B018.md", "in_short": "Gli antipattern mostrano come velocità apparente e delega cieca erodano affidabilità e apprendimento.", "summary": "Accettazione acritica, trappola del 70%, produzione senza verifica, contesto assente, batch eccessivi e misura teatrale formano una griglia diagnostica.", "limitation": "Il catalogo identifica rischi plausibili; non stima frequenza, severità o effetto causale degli antipattern.", "primary_theme": "T:Verification & review", "streams_json": "[\"T:Verification & review\", \"T:Code quality & security\", \"T:Skills & learning\", \"T:System knowledge & memory\", \"T:Productivity\"]", "topics_json": "[\"K:Review capacity and maintainers\", \"K:AI as reviewer\", \"K:Automation bias and trust\", \"K:Cognitive offloading and skill debt\", \"K:System understanding and rationale\", \"K:Ownership and knowledge transfer\", \"K:Controlled productivity evidence\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C02\", \"C:C08\", \"C:C10\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Antipatterns: What Not to Do\", \"line\": 5, \"offset\": 4, \"end\": 43}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Review capacity and maintainers", "K:AI as reviewer", "K:Automation bias and trust", "K:Cognitive offloading and skill debt", "K:System understanding and rationale", "K:Ownership and knowledge transfer", "K:Controlled productivity evidence", "K:Evaluation methodology and baselines" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter07/chapter07_antipatterns.tex" } }, { "id": "B019", "label": "Tools \\& Practices", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Productivity", "T:Skills & learning", "T:System knowledge & memory", "T:What agents can do" ], "actual": [], "mapped": [ "C:C03", "C:C09", "C:C11" ], "printed": false, "summary": "La guida agli strumenti collega scelta iniziale, contesto e pratiche operative.", "limitation": "La copertura tecnica è datata dall’edizione; i contenuti operativi appartengono alle schede dei file figli.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Tools \\& Practices", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10_v1.tex", "source_sha256": "2e995441dc5dc5d7fffdbf8664a422c202e117d9653b18298147e8d769b5d626", "line_count": 74, "word_count_approx": 309, "card_type": "chapter", "card_path": "cards/V01/B019.md", "in_short": "La guida agli strumenti collega scelta iniziale, contesto e pratiche operative.", "summary": "Il capitolo contenitore introduce dodici sezioni specializzate e spiega come comporre il percorso di implementazione.", "limitation": "La copertura tecnica è datata dall’edizione; i contenuti operativi appartengono alle schede dei file figli.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Productivity\", \"T:Skills & learning\", \"T:System knowledge & memory\", \"T:What agents can do\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Dependencies and agent supply chain\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C09\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"chapter\", \"title\": \"Tools \\\\& Practices\", \"line\": 2, \"offset\": 1, \"end\": 29}, {\"level\": \"subsection\", \"title\": \"The Journey Through This Chapter\", \"line\": 13, \"offset\": 417, \"end\": 463}, {\"level\": \"subsection\", \"title\": \"How to Read This Chapter\", \"line\": 32, \"offset\": 1981, \"end\": 2019}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Evaluation methodology and baselines", "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Dependencies and agent supply chain" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10_v1.tex" } }, { "id": "B020", "label": "Understanding Tool Maturity Levels", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Productivity", "T:Skills & learning", "T:Verification & review" ], "actual": [], "mapped": [ "C:C02", "C:C11" ], "printed": false, "summary": "I livelli di maturità aiutano a distinguere esplorazione, prova e adozione.", "limitation": "Le etichette ASSESS/TRIAL/ADOPT sono raccomandazioni contestuali e datate, non punteggi di affidabilità universali.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Understanding Tool Maturity Levels", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/01_maturity_levels.tex", "source_sha256": "03dbba47d7945250e812fd53c8e722015d520811b40ff0fd83c79aa2d63d3b21", "line_count": 151, "word_count_approx": 646, "card_type": "section", "card_path": "cards/V01/B020.md", "in_short": "I livelli di maturità aiutano a distinguere esplorazione, prova e adozione.", "summary": "Il file propone di interpretare classificazioni tipo Technology Radar e di far evolvere la scelta attraverso esperienza locale e revisione periodica.", "limitation": "Le etichette ASSESS/TRIAL/ADOPT sono raccomandazioni contestuali e datate, non punteggi di affidabilità universali.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Productivity\", \"T:Skills & learning\", \"T:Verification & review\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\", \"K:Controlled productivity evidence\"]", "editorial_chapters_json": "[\"C:C02\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Understanding Tool Maturity Levels\", \"line\": 3, \"offset\": 2, \"end\": 46}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Evaluation methodology and baselines", "K:Controlled productivity evidence" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/01_maturity_levels.tex" } }, { "id": "B021", "label": "Getting Started: Choosing Your Primary Tool", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Productivity", "T:Skills & learning", "T:What agents can do" ], "actual": [], "mapped": [ "C:C03", "C:C11" ], "printed": false, "summary": "La scelta dello strumento deve dipendere da workflow, vincoli e tipo di delega.", "limitation": "Caratteristiche, piani commerciali e raccomandazioni sono uno snapshot del testo; non sono state riverificate come catalogo attuale.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Getting Started: Choosing Your Primary Tool", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/02_choosing_tools.tex", "source_sha256": "fcaa3ca736559cad41090b0b9207ab7b4e84b5b41c5fba5051e1977110691750", "line_count": 363, "word_count_approx": 1993, "card_type": "section", "card_path": "cards/V01/B021.md", "in_short": "La scelta dello strumento deve dipendere da workflow, vincoli e tipo di delega.", "summary": "La sezione introduce il confronto fra strumenti principali e le informazioni che ciascuno vede, preparando la successiva progettazione del contesto.", "limitation": "Caratteristiche, piani commerciali e raccomandazioni sono uno snapshot del testo; non sono state riverificate come catalogo attuale.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Productivity\", \"T:Skills & learning\", \"T:What agents can do\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\", \"K:Agent failure and escalation\", \"K:Long-running agent execution\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Getting Started: Choosing Your Primary Tool\", \"line\": 2, \"offset\": 1, \"end\": 54}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Evaluation methodology and baselines", "K:Agent failure and escalation", "K:Long-running agent execution" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/02_choosing_tools.tex" } }, { "id": "B022", "label": "Static Context Engineering: Teaching AI About Your Project", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:System knowledge & memory", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C09" ], "printed": false, "summary": "Le istruzioni persistenti rendono esplicite convenzioni, vincoli e decisioni del progetto.", "limitation": "Un file di istruzioni può essere obsoleto o contraddittorio; la sua presenza non garantisce conformità del codice prodotto.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Static Context Engineering: Teaching AI About Your Project", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/03_static_context.tex", "source_sha256": "2fbca131864d7f4b1d1beeecea73b61e395b598486e47eac3176bfc9918f2b3a", "line_count": 424, "word_count_approx": 1782, "card_type": "section", "card_path": "cards/V01/B022.md", "in_short": "Le istruzioni persistenti rendono esplicite convenzioni, vincoli e decisioni del progetto.", "summary": "AGENTS.md e file analoghi sono trattati come una base di contesto mantenuta dal team, con esempi ed esercizi di configurazione.", "limitation": "Un file di istruzioni può essere obsoleto o contraddittorio; la sua presenza non garantisce conformità del codice prodotto.", "primary_theme": "T:System knowledge & memory", "streams_json": "[\"T:System knowledge & memory\", \"T:Engineering structure\"]", "topics_json": "[\"K:Agent context and persistent memory\", \"K:System understanding and rationale\"]", "editorial_chapters_json": "[\"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Static Context Engineering: Teaching AI About Your Project\", \"line\": 2, \"offset\": 1, \"end\": 69}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent context and persistent memory", "K:System understanding and rationale" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/03_static_context.tex" } }, { "id": "B023", "label": "Dynamic Context Engineering: Real-Time Information", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:System knowledge & memory", "T:What agents can do", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C06", "C:C09" ], "printed": false, "summary": "Il contesto dinamico recupera informazioni aggiornate nel momento in cui servono.", "limitation": "Freschezza, autorizzazione e attendibilità del contenuto devono essere verificate separatamente; un connettore può ampliare la superficie d’attacco.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Dynamic Context Engineering: Real-Time Information", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/04_dynamic_context.tex", "source_sha256": "68046ecb5f4a08a87ce5c30ad04826e3b08a84b9c57a5c648d7b3d2f3c40c61d", "line_count": 312, "word_count_approx": 1438, "card_type": "section", "card_path": "cards/V01/B023.md", "in_short": "Il contesto dinamico recupera informazioni aggiornate nel momento in cui servono.", "summary": "La sezione collega documentazione, schema dati e risorse esterne tramite connettori, distinguendoli dalle istruzioni statiche.", "limitation": "Freschezza, autorizzazione e attendibilità del contenuto devono essere verificate separatamente; un connettore può ampliare la superficie d’attacco.", "primary_theme": "T:System knowledge & memory", "streams_json": "[\"T:System knowledge & memory\", \"T:What agents can do\", \"T:Code quality & security\"]", "topics_json": "[\"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Dependencies and agent supply chain\", \"K:Prompt injection and malicious tools\"]", "editorial_chapters_json": "[\"C:C06\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Dynamic Context Engineering: Real-Time Information\", \"line\": 2, \"offset\": 1, \"end\": 61}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Dependencies and agent supply chain", "K:Prompt injection and malicious tools" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/04_dynamic_context.tex" } }, { "id": "B024", "label": "Context Flow Control: Managing Complex Sessions", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:System knowledge & memory", "T:What agents can do" ], "actual": [], "mapped": [ "C:C03", "C:C09" ], "printed": false, "summary": "Compattazione e handoff governano la continuità delle sessioni complesse.", "limitation": "Un riepilogo può perdere vincoli o spiegazioni; i passaggi di stato devono essere verificabili rispetto agli artefatti originari.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Context Flow Control: Managing Complex Sessions", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/05_context_flow.tex", "source_sha256": "e4c09a224cbac8339536664e6162fbafc708bae4e46667cf8008cd78f962f370", "line_count": 346, "word_count_approx": 1419, "card_type": "section", "card_path": "cards/V01/B024.md", "in_short": "Compattazione e handoff governano la continuità delle sessioni complesse.", "summary": "Plan mode, riepiloghi e passaggi strutturati sono strumenti per controllare l’accumulo del contesto e conservare lo stato del lavoro.", "limitation": "Un riepilogo può perdere vincoli o spiegazioni; i passaggi di stato devono essere verificabili rispetto agli artefatti originari.", "primary_theme": "T:System knowledge & memory", "streams_json": "[\"T:System knowledge & memory\", \"T:What agents can do\"]", "topics_json": "[\"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Long-running agent execution\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Context Flow Control: Managing Complex Sessions\", \"line\": 2, \"offset\": 1, \"end\": 58}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Long-running agent execution" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/05_context_flow.tex" } }, { "id": "B025", "label": "From Single Agent to Multi-Agent Workflows", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure", "T:Verification & review" ], "actual": [], "mapped": [ "C:C03", "C:C11" ], "printed": false, "summary": "Il passaggio a più agenti richiede specializzazione, confini e punti di controllo.", "limitation": "Il parallelismo non dimostra maggiore produttività; occorre contabilizzare errori di coordinamento e revisione seriale.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "From Single Agent to Multi-Agent Workflows", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/06_multi_agent.tex", "source_sha256": "0cdb4fb9d02a3fdf228a76b853e4d4c51db6999b0eb13acc4ad75626ad216bbd", "line_count": 487, "word_count_approx": 2714, "card_type": "section", "card_path": "cards/V01/B025.md", "in_short": "Il passaggio a più agenti richiede specializzazione, confini e punti di controllo.", "summary": "La sezione distingue subagent e workflow multi-agente, associando integrazione enterprise, condivisione del contesto e checkpoint umani.", "limitation": "Il parallelismo non dimostra maggiore produttività; occorre contabilizzare errori di coordinamento e revisione seriale.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Engineering structure\", \"T:Verification & review\"]", "topics_json": "[\"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"From Single Agent to Multi-Agent Workflows\", \"line\": 2, \"offset\": 1, \"end\": 53}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution", "K:Specifications and governance", "K:Ownership and knowledge transfer" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/06_multi_agent.tex" } }, { "id": "B026", "label": "Development Practices Enhanced by AI", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Verification & review", "T:Code quality & security", "T:System knowledge & memory", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C05", "C:C07", "C:C09" ], "printed": false, "summary": "TDD, comprensione del legacy e migrazioni sono casi concreti di lavoro assistito.", "limitation": "Le valutazioni di maturità sono contestuali; nelle migrazioni la preservazione semantica richiede casi indipendenti dal codice trasformato.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Development Practices Enhanced by AI", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/07_dev_practices.tex", "source_sha256": "d9b75f944c3c71265c0c686a6e658b4abe2baf222ee28eee7d41c2e3c6e1e7b0", "line_count": 816, "word_count_approx": 4234, "card_type": "section", "card_path": "cards/V01/B026.md", "in_short": "TDD, comprensione del legacy e migrazioni sono casi concreti di lavoro assistito.", "summary": "Il testo descrive come applicare AI a sviluppo guidato dai test, analisi del sistema, forward engineering e trasformazioni ripetitive.", "limitation": "Le valutazioni di maturità sono contestuali; nelle migrazioni la preservazione semantica richiede casi indipendenti dal codice trasformato.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Verification & review\", \"T:Code quality & security\", \"T:System knowledge & memory\", \"T:Skills & learning\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Evaluation methodology and baselines\", \"K:System understanding and rationale\", \"K:Onboarding and developer fluency\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C05\", \"C:C07\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Development Practices Enhanced by AI\", \"line\": 2, \"offset\": 1, \"end\": 47}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Specifications and governance", "K:Independent verification and oracles", "K:Evaluation methodology and baselines", "K:System understanding and rationale", "K:Onboarding and developer fluency", "K:Ownership and knowledge transfer" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/07_dev_practices.tex" } }, { "id": "B027", "label": "Team Collaboration \\& Knowledge Sharing", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Skills & learning", "T:System knowledge & memory", "T:Engineering structure", "T:Verification & review" ], "actual": [], "mapped": [ "C:C10", "C:C11" ], "printed": false, "summary": "L’adozione collettiva richiede trasmissione delle pratiche e decisioni condivise.", "limitation": "La condivisione degli artefatti non dimostra comprensione condivisa: servono verifiche di handoff e capacità autonome.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Team Collaboration \\& Knowledge Sharing", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/08_team_collaboration.tex", "source_sha256": "f25d348c036c0954b18c5d1837930f434106612fc32cf6ab1562a67bfd80db3e", "line_count": 330, "word_count_approx": 1073, "card_type": "section", "card_path": "cards/V01/B027.md", "in_short": "L’adozione collettiva richiede trasmissione delle pratiche e decisioni condivise.", "summary": "La sezione tratta librerie di prompt, collaborazione, competenze e ruoli complementari di persone e agenti.", "limitation": "La condivisione degli artefatti non dimostra comprensione condivisa: servono verifiche di handoff e capacità autonome.", "primary_theme": "T:Skills & learning", "streams_json": "[\"T:Skills & learning\", \"T:System knowledge & memory\", \"T:Engineering structure\", \"T:Verification & review\"]", "topics_json": "[\"K:Skills formation and teaching\", \"K:Cognitive offloading and skill debt\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C10\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Team Collaboration \\\\& Knowledge Sharing\", \"line\": 2, \"offset\": 1, \"end\": 50}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Skills formation and teaching", "K:Cognitive offloading and skill debt", "K:Specifications and governance", "K:Ownership and knowledge transfer" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/08_team_collaboration.tex" } }, { "id": "B028", "label": "Security, Compliance \\& Governance", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Code quality & security", "T:What agents can do", "T:Engineering structure", "T:System knowledge & memory", "T:Verification & review" ], "actual": [], "mapped": [ "C:C06", "C:C11" ], "printed": false, "summary": "La sicurezza degli agenti richiede difesa stratificata e controllo continuo.", "limitation": "Gli esempi non costituiscono un audit di conformità; distinguere regole applicate, osservabilità e rischio residuo.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Security, Compliance \\& Governance", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/09_security.tex", "source_sha256": "b8f5fff344bf416debc2851e29e9120b3ab90966db395ace2696f806ab0e4072", "line_count": 975, "word_count_approx": 4534, "card_type": "section", "card_path": "cards/V01/B028.md", "in_short": "La sicurezza degli agenti richiede difesa stratificata e controllo continuo.", "summary": "Policy-as-code, guardrail, flussi di dati, monitoraggio e risposta agli incidenti compongono il quadro operativo.", "limitation": "Gli esempi non costituiscono un audit di conformità; distinguere regole applicate, osservabilità e rischio residuo.", "primary_theme": "T:Code quality & security", "streams_json": "[\"T:Code quality & security\", \"T:What agents can do\", \"T:Engineering structure\", \"T:System knowledge & memory\", \"T:Verification & review\"]", "topics_json": "[\"K:Prompt injection and malicious tools\", \"K:Vulnerability and secure code\", \"K:Dependencies and agent supply chain\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C06\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Security, Compliance \\\\& Governance\", \"line\": 2, \"offset\": 1, \"end\": 45}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Prompt injection and malicious tools", "K:Vulnerability and secure code", "K:Dependencies and agent supply chain", "K:Specifications and governance", "K:Ownership and knowledge transfer" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/09_security.tex" } }, { "id": "B029", "label": "Measurement \\& Observability", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Productivity", "T:Economics", "T:Verification & review" ], "actual": [], "mapped": [ "C:C02", "C:C12" ], "printed": false, "summary": "Osservabilità e misura devono ricondurre i consumi al valore verificato.", "limitation": "Valore economico e percezione soggettiva hanno denominatori diversi; documentare baseline e campione per ogni confronto.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Measurement \\& Observability", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/10_measurement.tex", "source_sha256": "35163db617c256568b42f6a6377f5937b4880bf3c9726465623bcccc906b4f25", "line_count": 347, "word_count_approx": 1562, "card_type": "section", "card_path": "cards/V01/B029.md", "in_short": "Osservabilità e misura devono ricondurre i consumi al valore verificato.", "summary": "La sezione collega utilizzo, qualità, costi, debugging e flusso del lavoro, evitando la riduzione della produttività alle righe generate.", "limitation": "Valore economico e percezione soggettiva hanno denominatori diversi; documentare baseline e campione per ogni confronto.", "primary_theme": "T:Productivity", "streams_json": "[\"T:Productivity\", \"T:Economics\", \"T:Verification & review\"]", "topics_json": "[\"K:Output versus shipped value\", \"K:Evaluation methodology and baselines\", \"K:Controlled productivity evidence\", \"K:Economic costs and return\"]", "editorial_chapters_json": "[\"C:C02\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Measurement \\\\& Observability\", \"line\": 2, \"offset\": 1, \"end\": 39}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Output versus shipped value", "K:Evaluation methodology and baselines", "K:Controlled productivity evidence", "K:Economic costs and return" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/10_measurement.tex" } }, { "id": "B030", "label": "Common Pitfalls to Avoid", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Code quality & security", "T:Skills & learning", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C03", "C:C08", "C:C10" ], "printed": false, "summary": "Gli errori ricorrenti di adozione vanno diagnosticati prima di aggiungere strumenti.", "limitation": "La somiglianza fra antipattern non è una nuova osservazione indipendente; usarli come checklist e ipotesi da misurare.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Common Pitfalls to Avoid", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/11_pitfalls.tex", "source_sha256": "43b8d038a2bd2750d637c5b1b3b4304394d2b56c4b42d6af8e7ff4aa981be13e", "line_count": 1534, "word_count_approx": 5636, "card_type": "section", "card_path": "cards/V01/B030.md", "in_short": "Gli errori ricorrenti di adozione vanno diagnosticati prima di aggiungere strumenti.", "summary": "Il file approfondisce trappole pratiche e le raccorda agli antipattern del catalogo: contesto, batch, sicurezza, competenze e metriche.", "limitation": "La somiglianza fra antipattern non è una nuova osservazione indipendente; usarli come checklist e ipotesi da misurare.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Code quality & security\", \"T:Skills & learning\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Agent failure and escalation\", \"K:Evaluation methodology and baselines\", \"K:Review capacity and maintainers\", \"K:AI as reviewer\", \"K:Automation bias and trust\", \"K:Cognitive offloading and skill debt\", \"K:System understanding and rationale\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C08\", \"C:C10\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Common Pitfalls to Avoid\", \"line\": 2, \"offset\": 1, \"end\": 35}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent failure and escalation", "K:Evaluation methodology and baselines", "K:Review capacity and maintainers", "K:AI as reviewer", "K:Automation bias and trust", "K:Cognitive offloading and skill debt", "K:System understanding and rationale", "K:Ownership and knowledge transfer" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/11_pitfalls.tex" } }, { "id": "B031", "label": "First 90-Days Implementation Roadmap", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Productivity", "T:Skills & learning", "T:Verification & review" ], "actual": [], "mapped": [ "C:C02", "C:C11" ], "printed": false, "summary": "Un pilot di novanta giorni deve partire da baseline e concludersi con una decisione motivata.", "limitation": "Il calendario è prescrittivo; il successo va deciso su esiti e qualità misurati, non sul mero completamento delle settimane.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "First 90-Days Implementation Roadmap", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/12_roadmap.tex", "source_sha256": "eeb8597bff408474a39e7206e6f8bfa0cc022a5ed832d7092a94cb975e646d91", "line_count": 206, "word_count_approx": 1125, "card_type": "section", "card_path": "cards/V01/B031.md", "in_short": "Un pilot di novanta giorni deve partire da baseline e concludersi con una decisione motivata.", "summary": "La roadmap organizza preparazione, sperimentazione, passaggio fra gruppi e criteri di go/no-go, collegandosi alla roadmap strategica di sei mesi.", "limitation": "Il calendario è prescrittivo; il successo va deciso su esiti e qualità misurati, non sul mero completamento delle settimane.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Productivity\", \"T:Skills & learning\", \"T:Verification & review\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\", \"K:Controlled productivity evidence\"]", "editorial_chapters_json": "[\"C:C02\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"First 90-Days Implementation Roadmap\", \"line\": 2, \"offset\": 1, \"end\": 47}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Evaluation methodology and baselines", "K:Controlled productivity evidence" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/chapters_verbose/chapter10/12_roadmap.tex" } }, { "id": "B032", "label": "Quick Reference Guides", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Verification & review", "T:Code quality & security", "T:Productivity" ], "actual": [], "mapped": [ "C:C02", "C:C04", "C:C08" ], "printed": false, "summary": "Le schede rapide rendono utilizzabili criteri di controllo e misura durante il lavoro.", "limitation": "È un compendio del volume; non rafforza statisticamente le affermazioni ripetute né trasforma euristiche in percentuali misurate.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Quick Reference Guides", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/appendices/appendix_a.tex", "source_sha256": "a73043ba2d48b5b2b08823830033518d7658c45eef66034901dc4a7670e860c0", "line_count": 950, "word_count_approx": 2324, "card_type": "appendix", "card_path": "cards/V01/B032.md", "in_short": "Le schede rapide rendono utilizzabili criteri di controllo e misura durante il lavoro.", "summary": "L’appendice raccoglie checklist, domande di revisione, segnali di dipendenza dall’AI, criteri di policy e riferimenti pratici.", "limitation": "È un compendio del volume; non rafforza statisticamente le affermazioni ripetute né trasforma euristiche in percentuali misurate.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Verification & review\", \"T:Code quality & security\", \"T:Productivity\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Specifications and governance\", \"K:Review capacity and maintainers\", \"K:AI as reviewer\", \"K:Automation bias and trust\", \"K:Controlled productivity evidence\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C02\", \"C:C04\", \"C:C08\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The 70\\\\% Problem --- Quick Reference\", \"line\": 7, \"offset\": 241, \"end\": 287}, {\"level\": \"section\", \"title\": \"Warning Signs Your Team Is Over-Relying on AI\", \"line\": 184, \"offset\": 4362, \"end\": 4417}, {\"level\": \"section\", \"title\": \"Resources for Learning AI-Assisted Development\", \"line\": 481, \"offset\": 13402, \"end\": 13458}, {\"level\": \"section\", \"title\": \"Quick Reference Summary\", \"line\": 897, \"offset\": 27204, \"end\": 27237}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Specifications and governance", "K:Review capacity and maintainers", "K:AI as reviewer", "K:Automation bias and trust", "K:Controlled productivity evidence", "K:Evaluation methodology and baselines" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/appendices/appendix_a.tex" } }, { "id": "B033", "label": "Expert Insights", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Skills & learning", "T:System knowledge & memory", "T:Productivity", "T:Economics" ], "actual": [], "mapped": [ "C:C01", "C:C02", "C:C10" ], "printed": false, "summary": "Le testimonianze degli esperti sono organizzate come principi professionali.", "limitation": "L’attribuzione e il contesto delle citazioni richiedono fonte primaria puntuale; le testimonianze non sono prove causali.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Expert Insights", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/appendices/appendix_b1.tex", "source_sha256": "431de3883a50e3f4f912773b1806fb89615511e88c470f2beec94fdbdb56cab4", "line_count": 709, "word_count_approx": 2603, "card_type": "appendix", "card_path": "cards/V01/B033.md", "in_short": "Le testimonianze degli esperti sono organizzate come principi professionali.", "summary": "Il file aggrega citazioni e lezioni su test, costo, qualità, contesto e trasformazione della professione.", "limitation": "L’attribuzione e il contesto delle citazioni richiedono fonte primaria puntuale; le testimonianze non sono prove causali.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Skills & learning\", \"T:System knowledge & memory\", \"T:Productivity\", \"T:Economics\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Human-agent collaboration\", \"K:Skills formation and teaching\", \"K:Cognitive offloading and skill debt\", \"K:Output versus shipped value\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C02\", \"C:C10\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Expert Insights-Key Quotes \\\\& Principles\", \"line\": 12, \"offset\": 173, \"end\": 223}, {\"level\": \"section\", \"title\": \"Synthesized Wisdom: Key Principles\", \"line\": 574, \"offset\": 20713, \"end\": 20757}, {\"level\": \"section\", \"title\": \"Closing Thought\", \"line\": 679, \"offset\": 23647, \"end\": 23672}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Human-agent collaboration", "K:Skills formation and teaching", "K:Cognitive offloading and skill debt", "K:Output versus shipped value", "K:Evaluation methodology and baselines" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/appendices/appendix_b1.tex" } }, { "id": "B034", "label": "Implementation Guide and Solutions", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure", "T:Verification & review", "T:System knowledge & memory", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C03", "C:C07", "C:C09", "C:C11" ], "printed": false, "summary": "Le soluzioni operative traducono i workflow in profili di agenti, handoff ed esercizi.", "limitation": "Il codice e le configurazioni non sono stati eseguiti in questo passaggio; l’etichetta di verifica al dicembre 2025 resta un’affermazione del testo.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Implementation Guide and Solutions", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/appendices/appendix_c_v4.tex", "source_sha256": "7a56d65173ebada784e0909351e1dd174bda54e53bdf2d61981d30730fc2cd97", "line_count": 2199, "word_count_approx": 8680, "card_type": "appendix", "card_path": "cards/V01/B034.md", "in_short": "Le soluzioni operative traducono i workflow in profili di agenti, handoff ed esercizi.", "summary": "L’appendice contiene ruoli specializzati, schemi di payload, configurazioni MCP, subagent e soluzioni per sessioni isolate.", "limitation": "Il codice e le configurazioni non sono stati eseguiti in questo passaggio; l’etichetta di verifica al dicembre 2025 resta un’affermazione del testo.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Engineering structure\", \"T:Verification & review\", \"T:System knowledge & memory\", \"T:Code quality & security\"]", "topics_json": "[\"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Dependencies and agent supply chain\", \"K:Independent verification and oracles\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C07\", \"C:C09\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Implementing custom AI Agents\", \"line\": 14, \"offset\": 331, \"end\": 370}, {\"level\": \"section\", \"title\": \"Enterprise Workflow Agent Profiles - Chapter ch:new-workflows\", \"line\": 371, \"offset\": 19835, \"end\": 19912}, {\"level\": \"section\", \"title\": \"Handoff Payload Schema Reference\", \"line\": 1058, \"offset\": 38965, \"end\": 39007}, {\"level\": \"section\", \"title\": \"Exercise Solutions\", \"line\": 1183, \"offset\": 43849, \"end\": 43877}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution", "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Dependencies and agent supply chain", "K:Independent verification and oracles", "K:Evaluation methodology and baselines" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/appendices/appendix_c_v4.tex" } }, { "id": "B035", "label": "AI Adoption in the European Union", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Code quality & security", "T:Engineering structure", "T:Economics", "T:What agents can do" ], "actual": [], "mapped": [ "C:C06", "C:C11", "C:C12" ], "printed": false, "summary": "L’adozione europea combina responsabilità, provenienza e scelte di sovranità operativa.", "limitation": "Le affermazioni giuridiche e le scadenze sono solo indicizzate, non riconfermate: prima del riuso servono verifica normativa aggiornata e valutazione competente.", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "AI Adoption in the European Union", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V01", "volume_title": "Non-Deterministic Software Engineering", "source_path": "AI_SW_ENG/appendices/appendix_eu.tex", "source_sha256": "90feff4e44644d90c9be9a4fc45a98f5f2ddc32b60a0cd62089c01c14dd598ea", "line_count": 471, "word_count_approx": 3630, "card_type": "appendix", "card_path": "cards/V01/B035.md", "in_short": "L’adozione europea combina responsabilità, provenienza e scelte di sovranità operativa.", "summary": "L’appendice tratta quadro regolatorio, rischio delle dipendenze, dati, architetture ibride e AI-BOM come esigenze connesse alla pratica ingegneristica.", "limitation": "Le affermazioni giuridiche e le scadenze sono solo indicizzate, non riconfermate: prima del riuso servono verifica normativa aggiornata e valutazione competente.", "primary_theme": "T:Code quality & security", "streams_json": "[\"T:Code quality & security\", \"T:Engineering structure\", \"T:Economics\", \"T:What agents can do\"]", "topics_json": "[\"K:Dependencies and agent supply chain\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\", \"K:Prompt injection and malicious tools\", \"K:Vulnerability and secure code\", \"K:Economic costs and return\", \"K:Long-running agent execution\"]", "editorial_chapters_json": "[\"C:C06\", \"C:C11\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Introduction: A Different Calculus\", \"line\": 7, \"offset\": 239, \"end\": 283}, {\"level\": \"section\", \"title\": \"The Regulatory Landscape: Three Critical Frameworks\", \"line\": 53, \"offset\": 4016, \"end\": 4077}, {\"level\": \"section\", \"title\": \"Operational Sovereignty Assessment\", \"line\": 356, \"offset\": 23319, \"end\": 23363}, {\"level\": \"section\", \"title\": \"Conclusion: A European Path\", \"line\": 422, \"offset\": 26991, \"end\": 27028}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Dependencies and agent supply chain", "K:Specifications and governance", "K:Ownership and knowledge transfer", "K:Prompt injection and malicious tools", "K:Vulnerability and secure code", "K:Economic costs and return", "K:Long-running agent execution" ], "volumes": [ "V01" ], "book": { "volume": "V01", "title": "Non-Deterministic Software Engineering", "subtitle": "How to Build Reliable Software with AI Assistants Without Losing Quality, Security, or Control", "url": "https://github.com/popoloni/non_deterministic_sw_eng", "edition": "1", "source_path": "AI_SW_ENG/appendices/appendix_eu.tex" } }, { "id": "B036", "label": "Introduzioni e postfazione — The Bigger Picture", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:What agents can do" ], "actual": [], "mapped": [ "C:C01", "C:C04" ], "printed": false, "summary": "La postfazione riconduce SDD alla trasformazione più ampia dell’ingegneria.", "limitation": "Le note biografiche nello stesso file non sono usate come contributo tecnico o evidenza sul metodo.", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Introduzioni e postfazione — The Bigger Picture", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V02", "volume_title": "Non-Deterministic Spec-driven Development", "source_path": "AI_SDD_ENG/main.tex", "source_sha256": "92d0dbc28b96698b5f186b9c4b120330883575b5db0eacd4f1b2eba71e3339b4", "line_count": 352, "word_count_approx": 1690, "card_type": "structural_synthesis", "card_path": "cards/V02/B036.md", "in_short": "La postfazione riconduce SDD alla trasformazione più ampia dell’ingegneria.", "summary": "Il main include introduzioni alle parti e una chiusura concettuale, oltre ai rimandi al volume precedente. La scheda riguarda questa prosa autonoma, non la duplicazione dei capitoli inclusi.", "limitation": "Le note biografiche nello stesso file non sono usate come contributo tecnico o evidenza sul metodo.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:What agents can do\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Developer workflow and adoption\", \"K:Human-agent collaboration\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C04\"]", "evidence_anchors_json": "[{\"level\": \"chapter\", \"title\": \"Afterword: The Bigger Picture\", \"line\": 176, \"offset\": 6712, \"end\": 6752}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:Developer workflow and adoption", "K:Human-agent collaboration" ], "volumes": [ "V02" ], "book": { "volume": "V02", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "From Principles to Practice -- A Companion to Non-Deterministic Software Engineering", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "edition": "1", "source_path": "AI_SDD_ENG/main.tex" } }, { "id": "B037", "label": "Preface", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:Productivity", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C01", "C:C04" ], "printed": false, "summary": "La prefazione definisce pubblico, motivazione e rapporto con il volume fondativo.", "limitation": "È inquadramento editoriale; il rapporto fra i volumi non equivale a conferma indipendente delle tesi comuni.", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Preface", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V02", "volume_title": "Non-Deterministic Spec-driven Development", "source_path": "AI_SDD_ENG/frontmatter.tex", "source_sha256": "d05ef047f0eb9ac267123cc1d3222d4830d5af420c6b18101659c003c955161d", "line_count": 278, "word_count_approx": 1919, "card_type": "reading_guide", "card_path": "cards/V02/B037.md", "in_short": "La prefazione definisce pubblico, motivazione e rapporto con il volume fondativo.", "summary": "Il file presenta SDD come modo di rendere esplicito l’intento prima della generazione e chiarisce il percorso di lettura.", "limitation": "È inquadramento editoriale; il rapporto fra i volumi non equivale a conferma indipendente delle tesi comuni.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:Productivity\", \"T:Skills & learning\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C04\"]", "evidence_anchors_json": "[{\"level\": \"chapter\", \"title\": \"Preface\", \"line\": 96, \"offset\": 2127, \"end\": 2145}, {\"level\": \"subsection\", \"title\": \"Before You Begin\", \"line\": 144, \"offset\": 6289, \"end\": 6319}, {\"level\": \"subsubsection\", \"title\": \"Reading Paths by Role\", \"line\": 206, \"offset\": 9625, \"end\": 9663}, {\"level\": \"subsection\", \"title\": \"The Experts Who Shaped This Book\", \"line\": 257, \"offset\": 12207, \"end\": 12253}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:Developer workflow and adoption", "K:Evaluation methodology and baselines" ], "volumes": [ "V02" ], "book": { "volume": "V02", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "From Principles to Practice -- A Companion to Non-Deterministic Software Engineering", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "edition": "1", "source_path": "AI_SDD_ENG/frontmatter.tex" } }, { "id": "B038", "label": "The Shift", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review" ], "actual": [], "mapped": [ "C:C04", "C:C11" ], "printed": false, "summary": "L’intento diventa la fonte di riferimento del processo, con confini di autorità espliciti.", "limitation": "I principi sono prescrizioni progettuali, non una dimostrazione che ogni specifica sia completa o corretta.", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Shift", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V02", "volume_title": "Non-Deterministic Spec-driven Development", "source_path": "AI_SDD_ENG/chapters/chapter01_shift.tex", "source_sha256": "1a91b066014e9e6b3a16028d28873509535d11dc4643cdb4f4976d63d37187df", "line_count": 421, "word_count_approx": 3280, "card_type": "chapter", "card_path": "cards/V02/B038.md", "in_short": "L’intento diventa la fonte di riferimento del processo, con confini di autorità espliciti.", "summary": "Il capitolo oppone code-first e intent-first, formula cinque principi e introduce il sistema di confini a tre livelli.", "limitation": "I principi sono prescrizioni progettuali, non una dimostrazione che ogni specifica sia completa o corretta.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"From Code-First to Intent-First\", \"line\": 15, \"offset\": 529, \"end\": 570}, {\"level\": \"section\", \"title\": \"Vibe Coding vs.\\\\ Spec-Driven Development\", \"line\": 85, \"offset\": 5684, \"end\": 5734}, {\"level\": \"section\", \"title\": \"Five Principles That Actually Work\", \"line\": 142, \"offset\": 10759, \"end\": 10803}, {\"level\": \"section\", \"title\": \"The Three-Tier Boundary System\", \"line\": 260, \"offset\": 19455, \"end\": 19495}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:Ownership and knowledge transfer" ], "volumes": [ "V02" ], "book": { "volume": "V02", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "From Principles to Practice -- A Companion to Non-Deterministic Software Engineering", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "edition": "1", "source_path": "AI_SDD_ENG/chapters/chapter01_shift.tex" } }, { "id": "B039", "label": "The Process", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:Verification & review", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C04", "C:C07" ], "printed": false, "summary": "Quattro fasi e quattro gate organizzano il passaggio dall’intento all’implementazione.", "limitation": "La sequenza dei gate va distinta dalla rigidità waterfall: cambiamenti tardivi richiedono invalidazione e rivalidazione degli artefatti interessati.", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Process", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V02", "volume_title": "Non-Deterministic Spec-driven Development", "source_path": "AI_SDD_ENG/chapters/chapter02_process.tex", "source_sha256": "d496238586c77b97533a3851a22bb7a35ff1af1be938da000d1acd7b880781ce", "line_count": 557, "word_count_approx": 3520, "card_type": "chapter", "card_path": "cards/V02/B039.md", "in_short": "Quattro fasi e quattro gate organizzano il passaggio dall’intento all’implementazione.", "summary": "Costituzione, chiarimenti, selezione del modello, analisi di consistenza e checklist rendono esplicite le condizioni di avanzamento.", "limitation": "La sequenza dei gate va distinta dalla rigidità waterfall: cambiamenti tardivi richiedono invalidazione e rivalidazione degli artefatti interessati.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:Verification & review\", \"T:Code quality & security\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Independent verification and oracles\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C07\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Four Phases, Four Gates\", \"line\": 20, \"offset\": 1046, \"end\": 1079}, {\"level\": \"section\", \"title\": \"Model Selection Strategy\", \"line\": 241, \"offset\": 12809, \"end\": 12843}, {\"level\": \"section\", \"title\": \"The Clarification Phase\", \"line\": 361, \"offset\": 20337, \"end\": 20370}, {\"level\": \"section\", \"title\": \"Quality Checklists: Unit Tests for English\", \"line\": 467, \"offset\": 25412, \"end\": 25464}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Independent verification and oracles", "K:Evaluation methodology and baselines" ], "volumes": [ "V02" ], "book": { "volume": "V02", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "From Principles to Practice -- A Companion to Non-Deterministic Software Engineering", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "edition": "1", "source_path": "AI_SDD_ENG/chapters/chapter02_process.tex" } }, { "id": "B040", "label": "The Toolkit", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Verification & review" ], "actual": [], "mapped": [ "C:C03", "C:C04" ], "printed": false, "summary": "Spec Kit rende operativo il processo mediante comandi e artefatti.", "limitation": "Comandi e caratteristiche rappresentano la versione descritta; la loro correttezza attuale e l’esecuzione non sono state verificate qui.", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Toolkit", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V02", "volume_title": "Non-Deterministic Spec-driven Development", "source_path": "AI_SDD_ENG/chapters/chapter03_toolkit.tex", "source_sha256": "79e4d12f06ab8e1cc78d204aa3194b2037d409be96724e10c31bd2fe1efc8114", "line_count": 688, "word_count_approx": 3063, "card_type": "chapter", "card_path": "cards/V02/B040.md", "in_short": "Spec Kit rende operativo il processo mediante comandi e artefatti.", "summary": "Il capitolo introduce installazione, workflow, struttura dei file e un walkthrough, traducendo principi generali in una toolchain concreta.", "limitation": "Comandi e caratteristiche rappresentano la versione descritta; la loro correttezza attuale e l’esecuzione non sono state verificate qui.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Verification & review\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Specifications and governance\", \"K:Independent verification and oracles\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C04\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Introducing Spec Kit\", \"line\": 20, \"offset\": 954, \"end\": 984}, {\"level\": \"section\", \"title\": \"Installation and Setup\", \"line\": 80, \"offset\": 4022, \"end\": 4054}, {\"level\": \"section\", \"title\": \"The Artifact Structure\", \"line\": 301, \"offset\": 11627, \"end\": 11659}, {\"level\": \"section\", \"title\": \"Quick Walkthrough\", \"line\": 566, \"offset\": 22612, \"end\": 22639}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Specifications and governance", "K:Independent verification and oracles" ], "volumes": [ "V02" ], "book": { "volume": "V02", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "From Principles to Practice -- A Companion to Non-Deterministic Software Engineering", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "edition": "1", "source_path": "AI_SDD_ENG/chapters/chapter03_toolkit.tex" } }, { "id": "B041", "label": "The Integration", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:What agents can do", "T:Verification & review" ], "actual": [], "mapped": [ "C:C04", "C:C11" ], "printed": false, "summary": "SDD si integra con workflow esistenti e responsabilità dei team multi-ruolo.", "limitation": "La corrispondenza documentale non prova efficacia organizzativa; servono esperimenti su handoff, ambiguità e intenzioni modificate.", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Integration", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V02", "volume_title": "Non-Deterministic Spec-driven Development", "source_path": "AI_SDD_ENG/chapters/chapter04_integration.tex", "source_sha256": "c50eb677650ac6e7c616679dafbe4a864c4306cc720ccff9e355ec1cbec2fe84", "line_count": 2550, "word_count_approx": 9387, "card_type": "chapter", "card_path": "cards/V02/B041.md", "in_short": "SDD si integra con workflow esistenti e responsabilità dei team multi-ruolo.", "summary": "Le mappature verso i capitoli del primo volume e i playbook collegano specifiche, tooling e collaborazione in contesti enterprise e brownfield.", "limitation": "La corrispondenza documentale non prova efficacia organizzativa; servono esperimenti su handoff, ambiguità e intenzioni modificate.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:What agents can do\", \"T:Verification & review\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Mapping to Chapter 4 Workflows\", \"line\": 20, \"offset\": 969, \"end\": 1009}, {\"level\": \"section\", \"title\": \"Mapping to Chapter 10 Tooling\", \"line\": 96, \"offset\": 3619, \"end\": 3658}, {\"level\": \"section\", \"title\": \"SDD in Multi-Role Teams\", \"line\": 471, \"offset\": 17163, \"end\": 17196}, {\"level\": \"section\", \"title\": \"Use Case Playbooks\", \"line\": 1221, \"offset\": 44970, \"end\": 44998}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution", "K:Ownership and knowledge transfer" ], "volumes": [ "V02" ], "book": { "volume": "V02", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "From Principles to Practice -- A Companion to Non-Deterministic Software Engineering", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "edition": "1", "source_path": "AI_SDD_ENG/chapters/chapter04_integration.tex" } }, { "id": "B042", "label": "Advanced Patterns", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:System knowledge & memory", "T:What agents can do" ], "actual": [], "mapped": [ "C:C04", "C:C09", "C:C11" ], "printed": false, "summary": "Sette pattern estendono il catalogo originario alla gestione delle specifiche.", "limitation": "La ripetizione di un pattern nei volumi successivi non aumenta il numero di osservazioni; l’efficacia resta da valutare nel contesto d’uso.", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Advanced Patterns", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V02", "volume_title": "Non-Deterministic Spec-driven Development", "source_path": "AI_SDD_ENG/chapters/chapter05_patterns.tex", "source_sha256": "a5d81ef982282d03e93f32f344890d01e05eb138a89c84e7d4fda6818e602e0d", "line_count": 1107, "word_count_approx": 4763, "card_type": "chapter", "card_path": "cards/V02/B042.md", "in_short": "Sette pattern estendono il catalogo originario alla gestione delle specifiche.", "summary": "Costituzione, chiarificazione anticipata, consistenza, conservazione della ricerca, esplorazione parallela, scala delle specifiche e cambio del modello compongono un catalogo specifico.", "limitation": "La ripetizione di un pattern nei volumi successivi non aumenta il numero di osservazioni; l’efficacia resta da valutare nel contesto d’uso.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:System knowledge & memory\", \"T:What agents can do\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:System understanding and rationale\", \"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C09\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Extending the Pattern Catalog\", \"line\": 16, \"offset\": 494, \"end\": 534}, {\"level\": \"section\", \"title\": \"Pattern Overview\", \"line\": 85, \"offset\": 4218, \"end\": 4245}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:System understanding and rationale", "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution" ], "volumes": [ "V02" ], "book": { "volume": "V02", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "From Principles to Practice -- A Companion to Non-Deterministic Software Engineering", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "edition": "1", "source_path": "AI_SDD_ENG/chapters/chapter05_patterns.tex" } }, { "id": "B043", "label": "Limitations and Judgment", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:Productivity", "T:Verification & review" ], "actual": [], "mapped": [ "C:C02", "C:C04" ], "printed": false, "summary": "Il valore di SDD dipende dal contesto e dai costi di mantenimento degli artefatti.", "limitation": "Non presentare come misurate soglie di convenienza o riduzioni di difetti che il testo propone soltanto come euristiche.", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Limitations and Judgment", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V02", "volume_title": "Non-Deterministic Spec-driven Development", "source_path": "AI_SDD_ENG/chapters/chapter06_limitations.tex", "source_sha256": "7b1ea36d975dbe6f0affa38604fcf3948632188646a5489e4e404bb06222e48b", "line_count": 456, "word_count_approx": 2289, "card_type": "chapter", "card_path": "cards/V02/B043.md", "in_short": "Il valore di SDD dipende dal contesto e dai costi di mantenimento degli artefatti.", "summary": "Il capitolo affronta documenti troppo lunghi, casi inadatti, aggiornamento delle specifiche e misurazione del successo, ponendo limiti all’adozione indiscriminata.", "limitation": "Non presentare come misurate soglie di convenienza o riduzioni di difetti che il testo propone soltanto come euristiche.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:Productivity\", \"T:Verification & review\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Controlled productivity evidence\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C02\", \"C:C04\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Walls of Text Problem\", \"line\": 20, \"offset\": 974, \"end\": 1009}, {\"level\": \"section\", \"title\": \"When Spec-Driven Development Doesn't Fit\", \"line\": 138, \"offset\": 4680, \"end\": 4730}, {\"level\": \"section\", \"title\": \"Measuring SDD Success\", \"line\": 295, \"offset\": 10967, \"end\": 10998}, {\"level\": \"section\", \"title\": \"Honest Assessment\", \"line\": 379, \"offset\": 15955, \"end\": 15982}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Controlled productivity evidence", "K:Evaluation methodology and baselines" ], "volumes": [ "V02" ], "book": { "volume": "V02", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "From Principles to Practice -- A Companion to Non-Deterministic Software Engineering", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "edition": "1", "source_path": "AI_SDD_ENG/chapters/chapter06_limitations.tex" } }, { "id": "B044", "label": "Templates", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C04", "C:C09" ], "printed": false, "summary": "Template comuni rendono tracciabili costituzione, chiarimenti, consistenza e ricerca.", "limitation": "La compilazione di un template non garantisce correttezza, indipendenza o approvazione delle intenzioni contenute.", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Templates", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V02", "volume_title": "Non-Deterministic Spec-driven Development", "source_path": "AI_SDD_ENG/appendices/appendix_templates.tex", "source_sha256": "07b284ea9dddb099048b387e2a47520049633282bf799b0527b815f142da6968", "line_count": 352, "word_count_approx": 1012, "card_type": "appendix", "card_path": "cards/V02/B044.md", "in_short": "Template comuni rendono tracciabili costituzione, chiarimenti, consistenza e ricerca.", "summary": "L’appendice fornisce strutture riutilizzabili per gli artefatti centrali del processo, con indicazioni sul loro ruolo.", "limitation": "La compilazione di un template non garantisce correttezza, indipendenza o approvazione delle intenzioni contenute.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Template: Constitution\", \"line\": 10, \"offset\": 271, \"end\": 303}, {\"level\": \"section\", \"title\": \"Template: Clarification Log\", \"line\": 112, \"offset\": 2423, \"end\": 2460}, {\"level\": \"section\", \"title\": \"Template: Research Artifact\", \"line\": 269, \"offset\": 5939, \"end\": 5976}, {\"level\": \"section\", \"title\": \"Further Reading\", \"line\": 342, \"offset\": 7362, \"end\": 7387}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:Agent context and persistent memory", "K:System understanding and rationale" ], "volumes": [ "V02" ], "book": { "volume": "V02", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "From Principles to Practice -- A Companion to Non-Deterministic Software Engineering", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "edition": "1", "source_path": "AI_SDD_ENG/appendices/appendix_templates.tex" } }, { "id": "B045", "label": "Complete Implementation Examples", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:Verification & review", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C04", "C:C07" ], "printed": false, "summary": "Due esempi mostrano come gli artefatti si collegano in un percorso end-to-end.", "limitation": "Gli esempi sono didattici; non sono stati eseguiti né trasformati in pilot con esiti produttivi misurati.", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Complete Implementation Examples", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V02", "volume_title": "Non-Deterministic Spec-driven Development", "source_path": "AI_SDD_ENG/appendices/appendix_examples.tex", "source_sha256": "9e3081d62ba108f76aeb1189b308242273d9900bbe0ed9756e2ea979c0e2af26", "line_count": 762, "word_count_approx": 3039, "card_type": "appendix", "card_path": "cards/V02/B045.md", "in_short": "Due esempi mostrano come gli artefatti si collegano in un percorso end-to-end.", "summary": "Notifications N+1 e Sprintify rappresentano rispettivamente estensione di un sistema e greenfield enterprise; il contributo è la continuità fra specifica, pianificazione e verifica.", "limitation": "Gli esempi sono didattici; non sono stati eseguiti né trasformati in pilot con esiti produttivi misurati.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:Verification & review\", \"T:Code quality & security\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Independent verification and oracles\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C07\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Choosing the Right Example\", \"line\": 10, \"offset\": 375, \"end\": 411}, {\"level\": \"section\", \"title\": \"Example: Notifications Feature (N+1)\", \"line\": 38, \"offset\": 1597, \"end\": 1643}, {\"level\": \"section\", \"title\": \"Example: Sprintify (Enterprise Greenfield)\", \"line\": 215, \"offset\": 7275, \"end\": 7327}, {\"level\": \"section\", \"title\": \"Further Reading\", \"line\": 755, \"offset\": 25151, \"end\": 25176}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Independent verification and oracles", "K:Evaluation methodology and baselines" ], "volumes": [ "V02" ], "book": { "volume": "V02", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "From Principles to Practice -- A Companion to Non-Deterministic Software Engineering", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "edition": "1", "source_path": "AI_SDD_ENG/appendices/appendix_examples.tex" } }, { "id": "B046", "label": "Playbook Quick Reference", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:Productivity", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C04", "C:C11" ], "printed": false, "summary": "I playbook adattano il processo a sviluppo nuovo, brownfield, hotfix e collaborazione.", "limitation": "Sono viste operative del medesimo metodo, non sei validazioni empiriche indipendenti.", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Playbook Quick Reference", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V02", "volume_title": "Non-Deterministic Spec-driven Development", "source_path": "AI_SDD_ENG/appendices/appendix_playbooks.tex", "source_sha256": "cce6b0f819e0f68a30b481b80e52619afdede7e5b60caf4ceb279b6506d77939", "line_count": 503, "word_count_approx": 1741, "card_type": "appendix", "card_path": "cards/V02/B046.md", "in_short": "I playbook adattano il processo a sviluppo nuovo, brownfield, hotfix e collaborazione.", "summary": "Checklist e sequenze di comandi condensano i sei scenari del capitolo di integrazione, rendendo esplicito cosa varia con il rischio e il contesto.", "limitation": "Sono viste operative del medesimo metodo, non sei validazioni empiriche indipendenti.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:Productivity\", \"T:Skills & learning\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Playbook Selection Guide\", \"line\": 10, \"offset\": 365, \"end\": 399}, {\"level\": \"section\", \"title\": \"Legacy Codebase Integration\", \"line\": 139, \"offset\": 4390, \"end\": 4427}, {\"level\": \"section\", \"title\": \"Spike/Exploration\", \"line\": 269, \"offset\": 8283, \"end\": 8310}, {\"level\": \"section\", \"title\": \"Quick Decision Reference\", \"line\": 474, \"offset\": 14253, \"end\": 14287}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Developer workflow and adoption", "K:Evaluation methodology and baselines" ], "volumes": [ "V02" ], "book": { "volume": "V02", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "From Principles to Practice -- A Companion to Non-Deterministic Software Engineering", "url": "https://www.amazon.com/dp/B0GGZ44RNX", "edition": "1", "source_path": "AI_SDD_ENG/appendices/appendix_playbooks.tex" } }, { "id": "B047", "label": "Introduzioni alle parti — SDD, seconda edizione", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Verification & review" ], "actual": [], "mapped": [ "C:C01", "C:C04", "C:C11" ], "printed": false, "summary": "Le introduzioni alle parti spiegano il passaggio dal processo alla progettazione dei framework.", "limitation": "Il ruolo è strutturale e sintetico; non va contato come prova separata per tutte le affermazioni dei capitoli.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Introduzioni alle parti — SDD, seconda edizione", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/main.tex", "source_sha256": "b2dace7f741f9577eaa99ecafa3ea28ac72907aabf565bdaa46373d147daddda", "line_count": 273, "word_count_approx": 1566, "card_type": "structural_synthesis", "card_path": "cards/V03/B047.md", "in_short": "Le introduzioni alle parti spiegano il passaggio dal processo alla progettazione dei framework.", "summary": "Il main della seconda edizione orienta fra anatomia, ecosistema, confronto, Enterprise SDD, pattern e limiti. Mantiene leggibile l’architettura del volume senza duplicarne i file inclusi.", "limitation": "Il ruolo è strutturale e sintetico; non va contato come prova separata per tutte le affermazioni dei capitoli.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Verification & review\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Human-agent collaboration\", \"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"title\": \"Introduzioni alle parti — SDD, seconda edizione\", \"line\": 1}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Human-agent collaboration", "K:Specifications and governance", "K:Independent verification and oracles", "K:Ownership and knowledge transfer" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/main.tex" } }, { "id": "B048", "label": "Preface", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:Productivity", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C01", "C:C04" ], "printed": false, "summary": "La prefazione esplicita l’allargamento dalla singola toolchain all’ecosistema SDD.", "limitation": "L’essere seconda edizione non dimostra da solo validazione del metodo o superiorità delle nuove soluzioni.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Preface", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/frontmatter.tex", "source_sha256": "deaf09fefde8f6df71c5e1124d9bbf194d51b0f46b1545c19ed8fbcbeb10772f", "line_count": 175, "word_count_approx": 1821, "card_type": "reading_guide", "card_path": "cards/V03/B048.md", "in_short": "La prefazione esplicita l’allargamento dalla singola toolchain all’ecosistema SDD.", "summary": "Il file contestualizza obiettivi, pubblico e novità della seconda edizione, preparando una lettura comparativa e progettuale.", "limitation": "L’essere seconda edizione non dimostra da solo validazione del metodo o superiorità delle nuove soluzioni.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:Productivity\", \"T:Skills & learning\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C04\"]", "evidence_anchors_json": "[{\"level\": \"chapter\", \"title\": \"Preface\", \"line\": 57, \"offset\": 2540, \"end\": 2558}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:Developer workflow and adoption", "K:Evaluation methodology and baselines" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/frontmatter.tex" } }, { "id": "B049", "label": "The Shift", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review" ], "actual": [], "mapped": [ "C:C04", "C:C11" ], "printed": false, "summary": "La base intent-first viene mantenuta e applicata a PROSE.", "limitation": "La relazione con B038 è di revisione editoriale: non trattare le formulazioni comuni come evidenze indipendenti.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Shift", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter01_shift.tex", "source_sha256": "17398f66766c49f10a2ca3b49efd9e5ce34ec76a8ec4f7fa7c5eda80202770d3", "line_count": 480, "word_count_approx": 4214, "card_type": "chapter", "card_path": "cards/V03/B049.md", "in_short": "La base intent-first viene mantenuta e applicata a PROSE.", "summary": "Il capitolo riprende il cambiamento di fonte di verità, i cinque principi e i confini a tre livelli, con un’applicazione più esplicita rispetto alla prima edizione.", "limitation": "La relazione con B038 è di revisione editoriale: non trattare le formulazioni comuni come evidenze indipendenti.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"From Code-First to Intent-First\", \"line\": 49, \"offset\": 1949, \"end\": 1990}, {\"level\": \"section\", \"title\": \"Vibe Coding vs.\\\\ Spec-Driven Development\", \"line\": 123, \"offset\": 7607, \"end\": 7657}, {\"level\": \"section\", \"title\": \"Five Principles That Actually Work --- Applied to PROSE\", \"line\": 296, \"offset\": 23399, \"end\": 23464}, {\"level\": \"section\", \"title\": \"The Three-Tier Boundary System\", \"line\": 344, \"offset\": 28718, \"end\": 28758}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:Ownership and knowledge transfer" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter01_shift.tex" } }, { "id": "B050", "label": "The Process: Four Phases, Four Gates", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:Verification & review", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C04", "C:C07" ], "printed": false, "summary": "Il processo a gate viene generalizzato oltre una singola implementazione.", "limitation": "Un gate può essere superato con artefatti formalmente coerenti ma intenzioni sbagliate; definire cosa il controllo dimostra e cosa lascia aperto.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Process: Four Phases, Four Gates", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter02_process.tex", "source_sha256": "be96b58d8095aafc81871af38d2b2fbd0451e9a3427b85b44777293235de5ae8", "line_count": 328, "word_count_approx": 4123, "card_type": "chapter", "card_path": "cards/V03/B050.md", "in_short": "Il processo a gate viene generalizzato oltre una singola implementazione.", "summary": "La seconda edizione aggiunge refinement stack, selezione per fase, confronto fra istanziazioni e failure mode delle quattro fasi.", "limitation": "Un gate può essere superato con artefatti formalmente coerenti ma intenzioni sbagliate; definire cosa il controllo dimostra e cosa lascia aperto.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:Verification & review\", \"T:Code quality & security\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Independent verification and oracles\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C07\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Four Phases, Four Gates\", \"line\": 22, \"offset\": 1870, \"end\": 1903}, {\"level\": \"section\", \"title\": \"The Refinement Stack\", \"line\": 157, \"offset\": 14493, \"end\": 14523}, {\"level\": \"section\", \"title\": \"Consistency Analysis\", \"line\": 199, \"offset\": 18516, \"end\": 18546}, {\"level\": \"section\", \"title\": \"Common Failure Modes by Phase\", \"line\": 276, \"offset\": 27056, \"end\": 27095}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Independent verification and oracles", "K:Evaluation methodology and baselines" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter02_process.tex" } }, { "id": "B051", "label": "Anatomy of an AI SDD Framework", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:System knowledge & memory", "T:What agents can do", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C03", "C:C04", "C:C09" ], "printed": false, "summary": "I framework vengono scomposti in istruzioni, prompt, agenti e skill su diversi substrati.", "limitation": "La tassonomia è una sintesi progettuale dell’autore; la somiglianza delle strutture non implica identità di comportamento o qualità.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Anatomy of an AI SDD Framework", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter03_anatomy.tex", "source_sha256": "3c9e70a2ae890e4300cbed66e1c7e86559ecd40c0815ac9937d1e618b4320059", "line_count": 426, "word_count_approx": 5133, "card_type": "chapter", "card_path": "cards/V03/B051.md", "in_short": "I framework vengono scomposti in istruzioni, prompt, agenti e skill su diversi substrati.", "summary": "La tassonomia delle primitive e le regole di composizione permettono di confrontare implementazioni apparentemente diverse e di leggere le strutture dei repository.", "limitation": "La tassonomia è una sintesi progettuale dell’autore; la somiglianza delle strutture non implica identità di comportamento o qualità.", "primary_theme": "T:System knowledge & memory", "streams_json": "[\"T:System knowledge & memory\", \"T:What agents can do\", \"T:Engineering structure\"]", "topics_json": "[\"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Dependencies and agent supply chain\", \"K:Developer workflow and adoption\", \"K:Specifications and governance\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C04\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Why Frameworks Look Alike Underneath\", \"line\": 15, \"offset\": 919, \"end\": 965}, {\"level\": \"section\", \"title\": \"Primitive 3 --- Agents\", \"line\": 152, \"offset\": 11392, \"end\": 11424}, {\"level\": \"section\", \"title\": \"Where the Primitives Live in the Three Big Harnesses\", \"line\": 285, \"offset\": 22622, \"end\": 22684}, {\"level\": \"section\", \"title\": \"What to Look For When You Read Part II\", \"line\": 396, \"offset\": 33860, \"end\": 33908}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Dependencies and agent supply chain", "K:Developer workflow and adoption", "K:Specifications and governance" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter03_anatomy.tex" } }, { "id": "B052", "label": "The Ecosystem: A Map of Public Frameworks", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Productivity", "T:Skills & learning", "T:What agents can do" ], "actual": [], "mapped": [ "C:C03", "C:C04", "C:C11" ], "printed": false, "summary": "L’ecosistema viene organizzato in famiglie e scelte di progettazione.", "limitation": "I profili sono uno snapshot del corpus del libro; popolarità e presenza nella mappa non sono valutazioni sperimentali.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Ecosystem: A Map of Public Frameworks", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter04_ecosystem.tex", "source_sha256": "23468c124cf3c33c4f866c6a440041024e0d09b9f677a2f895167c9ad0d9d959", "line_count": 228, "word_count_approx": 2298, "card_type": "chapter", "card_path": "cards/V03/B052.md", "in_short": "L’ecosistema viene organizzato in famiglie e scelte di progettazione.", "summary": "Il capitolo presenta profili dei framework, una tassonomia e una tesi di convergenza, orientando il lettore prima dei confronti dettagliati.", "limitation": "I profili sono uno snapshot del corpus del libro; popolarità e presenza nella mappa non sono valutazioni sperimentali.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Productivity\", \"T:Skills & learning\", \"T:What agents can do\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\", \"K:Specifications and governance\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Why the Ecosystem Fragmented\", \"line\": 16, \"offset\": 721, \"end\": 759}, {\"level\": \"section\", \"title\": \"The Four Must-Know Frameworks\", \"line\": 112, \"offset\": 8847, \"end\": 8886}, {\"level\": \"section\", \"title\": \"Framework Profiles\", \"line\": 117, \"offset\": 9934, \"end\": 9962}, {\"level\": \"section\", \"title\": \"How to Read Part II\", \"line\": 196, \"offset\": 15137, \"end\": 15166}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Evaluation methodology and baselines", "K:Specifications and governance" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter04_ecosystem.tex" } }, { "id": "B053", "label": "The Lightweight Protocols: GSD v1 and GSD-2", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:What agents can do" ], "actual": [], "mapped": [ "C:C03", "C:C04" ], "printed": false, "summary": "GSD v1 e GSD-2 rappresentano due modi di ottenere disciplina con poco processo.", "limitation": "La field-test simulation resta una simulazione didattica, non un confronto controllato eseguito sui prodotti attuali.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Lightweight Protocols: GSD v1 and GSD-2", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter05_gsd.tex", "source_sha256": "9f1b1ebed1f3956520fc13d642ded6df5bc2479d436ca3a40b33e65cf724e1df", "line_count": 303, "word_count_approx": 3954, "card_type": "chapter", "card_path": "cards/V03/B053.md", "in_short": "GSD v1 e GSD-2 rappresentano due modi di ottenere disciplina con poco processo.", "summary": "Il confronto combina workflow leggeri, evoluzione strutturata e una simulazione di autenticazione JWT per rendere visibili artefatti e transizioni.", "limitation": "La field-test simulation resta una simulazione didattica, non un confronto controllato eseguito sui prodotti attuali.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:What agents can do\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Agent failure and escalation\", \"K:Long-running agent execution\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C04\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Lightweight Philosophy\", \"line\": 16, \"offset\": 403, \"end\": 439}, {\"level\": \"section\", \"title\": \"GSD v1 --- Action-Oriented Protocol\", \"line\": 33, \"offset\": 2928, \"end\": 2973}, {\"level\": \"section\", \"title\": \"Field-Test Simulation: JWT Authentication\", \"line\": 203, \"offset\": 23010, \"end\": 23061}, {\"level\": \"section\", \"title\": \"Lessons for Your Practice\", \"line\": 241, \"offset\": 25845, \"end\": 25880}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Agent failure and escalation", "K:Long-running agent execution" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter05_gsd.tex" } }, { "id": "B054", "label": "The Specification Toolkits: Spec Kit and OpenSpec (openspec.dev)", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:Verification & review" ], "actual": [], "mapped": [ "C:C04", "C:C05" ], "printed": false, "summary": "Spec Kit e OpenSpec affrontano specificazione e cambiamento brownfield con strutture diverse.", "limitation": "Comandi e lifecycle vanno letti nella versione fornita; il confronto non dimostra la superiorità generale di uno strumento.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Specification Toolkits: Spec Kit and OpenSpec (openspec.dev)", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter06_spectools.tex", "source_sha256": "a803100f32d0ed9bb03506561afe3e44e8e838131df018e142c312206c9c9ea2", "line_count": 814, "word_count_approx": 10419, "card_type": "chapter", "card_path": "cards/V03/B054.md", "in_short": "Spec Kit e OpenSpec affrontano specificazione e cambiamento brownfield con strutture diverse.", "summary": "Il capitolo compara toolchain, delta specification, chiarimento prima del piano e una simulazione JWT, esplicitando trade-off e scelte di design.", "limitation": "Comandi e lifecycle vanno letti nella versione fornita; il confronto non dimostra la superiorità generale di uno strumento.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:Verification & review\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Independent verification and oracles\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C05\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Spec Kit --- The Specification Toolchain\", \"line\": 23, \"offset\": 1281, \"end\": 1331}, {\"level\": \"section\", \"title\": \"Spec Kit vs.\\\\ OpenSpec: A Direct Comparison\", \"line\": 496, \"offset\": 54876, \"end\": 54929}, {\"level\": \"section\", \"title\": \"The Clarify-Before-Plan Pattern\", \"line\": 585, \"offset\": 63086, \"end\": 63127}, {\"level\": \"section\", \"title\": \"Lessons for Your Practice\", \"line\": 766, \"offset\": 77336, \"end\": 77371}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Independent verification and oracles" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter06_spectools.tex" } }, { "id": "B055", "label": "The Agentic Orchestrators: BMAD Method and AI-RPI Protocol", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure", "T:Verification & review", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C03", "C:C04", "C:C11" ], "printed": false, "summary": "BMAD e AI-RPI mostrano due filosofie di coordinamento e handoff.", "limitation": "La specializzazione nominale degli agenti non assicura indipendenza epistemica; controllare le dipendenze fra input, artefatti e verifiche.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Agentic Orchestrators: BMAD Method and AI-RPI Protocol", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter07_orchestrators.tex", "source_sha256": "26356810d67207d7db408d49a5678729fc25eb2a0d104e88c964dd48334b2760", "line_count": 586, "word_count_approx": 7585, "card_type": "chapter", "card_path": "cards/V03/B055.md", "in_short": "BMAD e AI-RPI mostrano due filosofie di coordinamento e handoff.", "summary": "Il capitolo distingue persona e ruolo, orchestrazione agile e protocollo Research–Plan–Implement, usando una simulazione di progettazione di microservizio.", "limitation": "La specializzazione nominale degli agenti non assicura indipendenza epistemica; controllare le dipendenze fra input, artefatti e verifiche.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Engineering structure\", \"T:Verification & review\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\", \"K:Specifications and governance\", \"K:System understanding and rationale\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Orchestration Problem\", \"line\": 20, \"offset\": 946, \"end\": 981}, {\"level\": \"section\", \"title\": \"AI-RPI Protocol --- Adaptive Research--Plan--Implement Discipline\", \"line\": 322, \"offset\": 29166, \"end\": 29241}, {\"level\": \"section\", \"title\": \"The Handoff Problem\", \"line\": 422, \"offset\": 45892, \"end\": 45921}, {\"level\": \"section\", \"title\": \"Lessons for Your Practice\", \"line\": 487, \"offset\": 51099, \"end\": 51134}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution", "K:Specifications and governance", "K:System understanding and rationale" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter07_orchestrators.tex" } }, { "id": "B056", "label": "Autonomous Loops and Behavioral Foundations", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C03", "C:C04", "C:C07" ], "printed": false, "summary": "Loop autonomi e disciplina comportamentale sono componenti da comporre, non alternative equivalenti.", "limitation": "I profili e la simulazione non costituiscono un benchmark; distinguere regole suggerite e controlli realmente applicati dal runtime.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Autonomous Loops and Behavioral Foundations", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter08_autonomous.tex", "source_sha256": "5bff2e88df8bb7940b34d377d69d9b488bb0ba5100b4278946676015b0e011df", "line_count": 629, "word_count_approx": 10585, "card_type": "chapter", "card_path": "cards/V03/B056.md", "in_short": "Loop autonomi e disciplina comportamentale sono componenti da comporre, non alternative equivalenti.", "summary": "Ralph Copilot, Spec Loop Engine, skill ispirate a Karpathy, Agent Skills e AIDD sono letti attraverso autonomia, vincoli e una simulazione di implementazione.", "limitation": "I profili e la simulazione non costituiscono un benchmark; distinguere regole suggerite e controlli realmente applicati dal runtime.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Engineering structure\"]", "topics_json": "[\"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\", \"K:Developer workflow and adoption\", \"K:Specifications and governance\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C04\", \"C:C07\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Autonomy Spectrum\", \"line\": 20, \"offset\": 1167, \"end\": 1198}, {\"level\": \"section\", \"title\": \"Karpathy-Inspired Skills --- Behavioral Discipline\", \"line\": 257, \"offset\": 34923, \"end\": 34983}, {\"level\": \"section\", \"title\": \"AIDD --- Architecture-Decision-Driven Backend Pipeline\", \"line\": 426, \"offset\": 52965, \"end\": 53029}, {\"level\": \"section\", \"title\": \"Lessons for Your Practice\", \"line\": 581, \"offset\": 73480, \"end\": 73515}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles", "K:Developer workflow and adoption", "K:Specifications and governance" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter08_autonomous.tex" } }, { "id": "B057", "label": "The Comparison: Patterns, Divergences, and Convergences", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Productivity", "T:Skills & learning", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C03", "C:C04", "C:C11" ], "printed": false, "summary": "Dodici dimensioni rendono espliciti convergenze e disaccordi fra framework.", "limitation": "Le posizioni nelle mappe e nei quadranti sono giudizi editoriali, non misure statistiche o classifiche indipendenti.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Comparison: Patterns, Divergences, and Convergences", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter09_comparison.tex", "source_sha256": "45be338524ea5b8bf299ba0d524c0959fbadcdc38bdd0a5b81d6fa3dfad45262", "line_count": 626, "word_count_approx": 8041, "card_type": "chapter", "card_path": "cards/V03/B057.md", "in_short": "Dodici dimensioni rendono espliciti convergenze e disaccordi fra framework.", "summary": "Tabella di confronto, mappe strategiche e quattro divergenze filosofiche organizzano le scelte senza limitarsi a elencare funzionalità.", "limitation": "Le posizioni nelle mappe e nei quadranti sono giudizi editoriali, non misure statistiche o classifiche indipendenti.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Productivity\", \"T:Skills & learning\", \"T:Code quality & security\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\", \"K:Architecture and modularity\", \"K:Technical debt and complexity\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Twelve Core Comparison Dimensions\", \"line\": 20, \"offset\": 1022, \"end\": 1069}, {\"level\": \"section\", \"title\": \"Strategic Maps: Three Magic Quadrants\", \"line\": 171, \"offset\": 18678, \"end\": 18725}, {\"level\": \"section\", \"title\": \"Thematic Deep-Dives\", \"line\": 429, \"offset\": 34076, \"end\": 34105}, {\"level\": \"section\", \"title\": \"The Divergence Map: Four Irreducible Philosophical Splits\", \"line\": 575, \"offset\": 58472, \"end\": 58539}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Evaluation methodology and baselines", "K:Architecture and modularity", "K:Technical debt and complexity" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter09_comparison.tex" } }, { "id": "B058", "label": "Design Boundaries: What to Adopt, What to Avoid", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Productivity", "T:Skills & learning", "T:Verification & review" ], "actual": [], "mapped": [ "C:C04", "C:C11" ], "printed": false, "summary": "Il framework adatto è quello con rigore sufficiente e costo proporzionato.", "limitation": "La selezione va verificata su rischi e vincoli locali; la tassonomia non rende equivalenti strumenti complementari.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Design Boundaries: What to Adopt, What to Avoid", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter10_boundaries.tex", "source_sha256": "f4e89dfab7a1b75f618433de0d8423b30f82d62a3d6994071abcf7352c46d135", "line_count": 403, "word_count_approx": 5427, "card_type": "chapter", "card_path": "cards/V03/B058.md", "in_short": "Il framework adatto è quello con rigore sufficiente e costo proporzionato.", "summary": "Tre livelli di rigore, griglia di adozione, antipattern, migrazione e complementarità definiscono confini per evitare accumulo indiscriminato di meccanismi.", "limitation": "La selezione va verificata su rischi e vincoli locali; la tassonomia non rende equivalenti strumenti complementari.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Productivity\", \"T:Skills & learning\", \"T:Verification & review\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Three Tiers of Rigor\", \"line\": 20, \"offset\": 756, \"end\": 790}, {\"level\": \"section\", \"title\": \"Eight Anti-Patterns from the Wild\", \"line\": 198, \"offset\": 19317, \"end\": 19360}, {\"level\": \"section\", \"title\": \"What an Ideal Framework Would Combine\", \"line\": 304, \"offset\": 29330, \"end\": 29377}, {\"level\": \"section\", \"title\": \"Design Boundaries of Enterprise SDD\", \"line\": 363, \"offset\": 36495, \"end\": 36540}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Evaluation methodology and baselines", "K:Specifications and governance", "K:Ownership and knowledge transfer" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter10_boundaries.tex" } }, { "id": "B059", "label": "Constitution and Architecture: The Framework Foundation", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C04", "C:C05", "C:C11" ], "printed": false, "summary": "Enterprise SDD combina costituzione, architettura e progressione controllata.", "limitation": "La storia di composizione non è una validazione indipendente; mantenere separate scelte progettuali, conformità formale ed esiti reali.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Constitution and Architecture: The Framework Foundation", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter11_foundation.tex", "source_sha256": "8fadfb701661d64f294a2b51180d2fc1749af952a6f15c9e099ce3b2d4f6b4fc", "line_count": 592, "word_count_approx": 4983, "card_type": "chapter", "card_path": "cards/V03/B059.md", "in_short": "Enterprise SDD combina costituzione, architettura e progressione controllata.", "summary": "Il capitolo racconta la composizione del framework attraverso vincoli inviolabili, pipeline a sei fasi e gate, raccordando l’ecosistema alla proposta dell’autore.", "limitation": "La storia di composizione non è una validazione indipendente; mantenere separate scelte progettuali, conformità formale ed esiti reali.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:Code quality & security\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Ownership and knowledge transfer\", \"K:Architecture and modularity\", \"K:Technical debt and complexity\", \"K:Independent verification and oracles\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C05\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Five Days With Enterprise SDD\", \"line\": 28, \"offset\": 1394, \"end\": 1433}, {\"level\": \"section\", \"title\": \"Core Philosophy\", \"line\": 139, \"offset\": 13030, \"end\": 13055}, {\"level\": \"section\", \"title\": \"The Six-Phase Pipeline\", \"line\": 183, \"offset\": 18800, \"end\": 18832}, {\"level\": \"section\", \"title\": \"Constitution-Driven Development in Practice\", \"line\": 414, \"offset\": 30897, \"end\": 30950}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Ownership and knowledge transfer", "K:Architecture and modularity", "K:Technical debt and complexity", "K:Independent verification and oracles" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter11_foundation.tex" } }, { "id": "B060", "label": "The Agent Pipeline and Gates", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure", "T:Verification & review", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C04", "C:C07", "C:C11" ], "printed": false, "summary": "La pipeline degli agenti rende espliciti artefatti, handoff e autorità dei gate.", "limitation": "Il numero di agenti e di gate non misura qualità; servono prove su fallimenti concorrenti, risposte tardive e intenzioni modificate.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Agent Pipeline and Gates", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter12_agents.tex", "source_sha256": "0eeb7ee0be66f6453866402f554996d7834eabd8d3648fa4e594b9c24442c901", "line_count": 867, "word_count_approx": 8254, "card_type": "chapter", "card_path": "cards/V03/B060.md", "in_short": "La pipeline degli agenti rende espliciti artefatti, handoff e autorità dei gate.", "summary": "Catalogo dei ruoli, agenti multi-mode, Three Amigos, livelli di cerimonia, pipeline di sicurezza e gestione delle domande descrivono il funzionamento interno di Enterprise SDD.", "limitation": "Il numero di agenti e di gate non misura qualità; servono prove su fallimenti concorrenti, risposte tardive e intenzioni modificate.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Engineering structure\", \"T:Verification & review\", \"T:Code quality & security\"]", "topics_json": "[\"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\", \"K:Independent verification and oracles\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C07\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Agent Topology at a Glance\", \"line\": 28, \"offset\": 1369, \"end\": 1409}, {\"level\": \"section\", \"title\": \"Complete Agent Catalog\", \"line\": 259, \"offset\": 21141, \"end\": 21173}, {\"level\": \"section\", \"title\": \"Gate Anatomy\", \"line\": 535, \"offset\": 43963, \"end\": 43985}, {\"level\": \"section\", \"title\": \"From sdd new to sdd ship : a full lap\", \"line\": 774, \"offset\": 58707, \"end\": 58767}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution", "K:Specifications and governance", "K:Ownership and knowledge transfer", "K:Independent verification and oracles", "K:Evaluation methodology and baselines" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter12_agents.tex" } }, { "id": "B061", "label": "Automation and Extensibility: CLI, Modules, and MCP", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:System knowledge & memory", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C03", "C:C06", "C:C09" ], "printed": false, "summary": "CLI, adattatori e moduli rendono estensibile il framework senza cambiare il suo nucleo concettuale.", "limitation": "Le superfici CLI e gli adattatori non sono stati eseguiti; compatibilità e sicurezza devono essere provate sulla versione concreta.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Automation and Extensibility: CLI, Modules, and MCP", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter13_automation.tex", "source_sha256": "e1f633e1cc1dca1637da2923b8c5275dc4707462a8677259629a68a52e023395", "line_count": 732, "word_count_approx": 6550, "card_type": "chapter", "card_path": "cards/V03/B061.md", "in_short": "CLI, adattatori e moduli rendono estensibile il framework senza cambiare il suo nucleo concettuale.", "summary": "Il capitolo presenta sessione terminale, preset, MCP, issue sync, skill e controlli di salute, mostrando come collegare agenti e artefatti all’ambiente operativo.", "limitation": "Le superfici CLI e gli adattatori non sono stati eseguiti; compatibilità e sicurezza devono essere provate sulla versione concreta.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:System knowledge & memory\", \"T:Code quality & security\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Specifications and governance\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Dependencies and agent supply chain\", \"K:Prompt injection and malicious tools\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C06\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"A 90-Second Terminal Session\", \"line\": 28, \"offset\": 1271, \"end\": 1309}, {\"level\": \"section\", \"title\": \"User Modules\", \"line\": 331, \"offset\": 23732, \"end\": 23754}, {\"level\": \"section\", \"title\": \"MCP Servers\", \"line\": 472, \"offset\": 32976, \"end\": 32997}, {\"level\": \"section\", \"title\": \"The sdd doctor Framework Health Check\", \"line\": 631, \"offset\": 43994, \"end\": 44048}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Specifications and governance", "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Dependencies and agent supply chain", "K:Prompt injection and malicious tools" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter13_automation.tex" } }, { "id": "B062", "label": "Autonomous Execution: Modes, Budgets, and Governance", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure", "T:Verification & review", "T:Economics", "T:Productivity" ], "actual": [], "mapped": [ "C:C03", "C:C11", "C:C12" ], "printed": false, "summary": "L’autonomia viene trattata come policy di esecuzione con budget ed escalation.", "limitation": "Budget e audit-after non sostituiscono criteri di accettazione indipendenti; gli esempi non provano sicurezza in produzione.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Autonomous Execution: Modes, Budgets, and Governance", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter14_autonomy.tex", "source_sha256": "484cf12509a4ef84dc7b2cb0174d8248a02443f033596e8667c04a8c7ea944dc", "line_count": 748, "word_count_approx": 7795, "card_type": "chapter", "card_path": "cards/V03/B062.md", "in_short": "L’autonomia viene trattata come policy di esecuzione con budget ed escalation.", "summary": "Tre modalità, un elemento per ciclo, evidence pack, convergence review e casi enterprise definiscono cosa l’agente può fare e quando deve fermarsi.", "limitation": "Budget e audit-after non sostituiscono criteri di accettazione indipendenti; gli esempi non provano sicurezza in produzione.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Engineering structure\", \"T:Verification & review\", \"T:Economics\", \"T:Productivity\"]", "topics_json": "[\"K:Agent failure and escalation\", \"K:Long-running agent execution\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\", \"K:Economic costs and return\", \"K:Output versus shipped value\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C11\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Use Case 1: The Friday Afternoon Decision\", \"line\": 28, \"offset\": 1445, \"end\": 1496}, {\"level\": \"section\", \"title\": \"The One-Item-Per-Cycle Protocol\", \"line\": 231, \"offset\": 16911, \"end\": 16952}, {\"level\": \"section\", \"title\": \"Convergence Review\", \"line\": 528, \"offset\": 37242, \"end\": 37270}, {\"level\": \"section\", \"title\": \"Use Case 2: Greenfield Enterprise Java with Kafka\", \"line\": 684, \"offset\": 49241, \"end\": 49300}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent failure and escalation", "K:Long-running agent execution", "K:Specifications and governance", "K:Ownership and knowledge transfer", "K:Economic costs and return", "K:Output versus shipped value" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter14_autonomy.tex" } }, { "id": "B063", "label": "Advanced Patterns: From Frameworks to Practice", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:What agents can do", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C04", "C:C06", "C:C07", "C:C11" ], "printed": false, "summary": "Diciannove pattern compongono gestione delle specifiche, coordinamento, sicurezza e autonomia.", "limitation": "I pattern condividono fonti e meccanismi con capitoli precedenti; non aggregarli come risultati autonomi di efficacia.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Advanced Patterns: From Frameworks to Practice", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter15_patterns.tex", "source_sha256": "4830546709bc5f2eb4641c47add3bb9992eb9089497514ac25f321b33f5f92aa", "line_count": 778, "word_count_approx": 11299, "card_type": "chapter", "card_path": "cards/V03/B063.md", "in_short": "Diciannove pattern compongono gestione delle specifiche, coordinamento, sicurezza e autonomia.", "summary": "I gruppi A–D includono chiarimento, red teaming, gap closure, handoff, anti-sycophancy, gate, supply chain, evidence pack e budget.", "limitation": "I pattern condividono fonti e meccanismi con capitoli precedenti; non aggregarli come risultati autonomi di efficacia.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:What agents can do\", \"T:Code quality & security\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\", \"K:Prompt injection and malicious tools\", \"K:Vulnerability and secure code\", \"K:Agent failure and escalation\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C06\", \"C:C07\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Pattern Structure\", \"line\": 20, \"offset\": 1015, \"end\": 1043}, {\"level\": \"section\", \"title\": \"Group B: Agent Coordination\", \"line\": 263, \"offset\": 25351, \"end\": 25388}, {\"level\": \"section\", \"title\": \"Group C: Quality and Safety\", \"line\": 396, \"offset\": 40080, \"end\": 40117}, {\"level\": \"section\", \"title\": \"Enterprise SDD Implementation Notes\", \"line\": 715, \"offset\": 75875, \"end\": 75920}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution", "K:Prompt injection and malicious tools", "K:Vulnerability and secure code", "K:Agent failure and escalation" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter15_patterns.tex" } }, { "id": "B064", "label": "Limitations and Professional Judgment", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:Productivity", "T:Verification & review", "T:What agents can do" ], "actual": [], "mapped": [ "C:C02", "C:C04", "C:C11" ], "printed": false, "summary": "Il giudizio professionale stabilisce quando ridurre, modificare o non usare SDD.", "limitation": "Witboost e Mollie sono resoconti del corpus, non pilot replicati qui; confronto e generalizzazione richiedono protocolli e denominatori.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Limitations and Professional Judgment", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter16_limitations.tex", "source_sha256": "c726789bc12ceb26cbed18d715409013fc83a4e244e8e90615cb413e05ee2ea6", "line_count": 542, "word_count_approx": 7471, "card_type": "chapter", "card_path": "cards/V03/B064.md", "in_short": "Il giudizio professionale stabilisce quando ridurre, modificare o non usare SDD.", "summary": "Rapporti di campo, anti-casi, overhead, gate skipping, spec rot e attriti organizzativi delimitano la portata del metodo.", "limitation": "Witboost e Mollie sono resoconti del corpus, non pilot replicati qui; confronto e generalizzazione richiedono protocolli e denominatori.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:Productivity\", \"T:Verification & review\", \"T:What agents can do\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Controlled productivity evidence\", \"K:Evaluation methodology and baselines\", \"K:Agent failure and escalation\"]", "editorial_chapters_json": "[\"C:C02\", \"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Two Field Reports: Witboost and Mollie\", \"line\": 20, \"offset\": 705, \"end\": 753}, {\"level\": \"section\", \"title\": \"The Overhead Tax\", \"line\": 184, \"offset\": 16802, \"end\": 16828}, {\"level\": \"section\", \"title\": \"The AI Reliability Gap\", \"line\": 335, \"offset\": 31115, \"end\": 31147}, {\"level\": \"section\", \"title\": \"The Professional Developer's Heuristic\", \"line\": 454, \"offset\": 44673, \"end\": 44721}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Controlled productivity evidence", "K:Evaluation methodology and baselines", "K:Agent failure and escalation" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter16_limitations.tex" } }, { "id": "B065", "label": "How to Design a New Framework", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Verification & review" ], "actual": [], "mapped": [ "C:C04", "C:C11" ], "printed": false, "summary": "Progettare un nuovo framework significa partire da problemi e vincoli osservati.", "limitation": "Il metodo di progettazione è una proposta; non va presentato come algoritmo che garantisce un framework più efficace.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "How to Design a New Framework", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/chapters/chapter17_design_framework.tex", "source_sha256": "995ee712004747b56e01abb6e71919950669b2896eba76087f46d27b86045fb9", "line_count": 325, "word_count_approx": 4897, "card_type": "chapter", "card_path": "cards/V03/B065.md", "in_short": "Progettare un nuovo framework significa partire da problemi e vincoli osservati.", "summary": "Four-Question Test, light evolution loop, storia di composizione di Enterprise SDD ed esempio Genesis definiscono una metodologia di evoluzione incrementale.", "limitation": "Il metodo di progettazione è una proposta; non va presentato come algoritmo che garantisce un framework più efficace.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Verification & review\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"A Light Methodology: The Four-Question Test\", \"line\": 12, \"offset\": 1059, \"end\": 1112}, {\"level\": \"section\", \"title\": \"The Light Evolution Loop\", \"line\": 99, \"offset\": 8450, \"end\": 8484}, {\"level\": \"section\", \"title\": \"Worked Example: Enterprise SDD's Composition Story\", \"line\": 115, \"offset\": 11200, \"end\": 11260}, {\"level\": \"section\", \"title\": \"Supporting Exemplar: Genesis as a Meta-Framework\", \"line\": 200, \"offset\": 23726, \"end\": 23784}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Specifications and governance", "K:Ownership and knowledge transfer" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/chapters/chapter17_design_framework.tex" } }, { "id": "B066", "label": "Installation and Setup", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Productivity", "T:Skills & learning", "T:Code quality & security", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C03", "C:C06" ], "printed": false, "summary": "La guida di installazione rende verificabili prerequisiti e configurazione del framework.", "limitation": "L’installazione non è stata ripetuta in questa analisi; richiede una verifica su ambiente, release e permessi effettivi.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Installation and Setup", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/appendix_setup.tex", "source_sha256": "6f4e96375b8e453691a7331cc62d4a78842d9323b43f9956152f32e5a2ff6ff3", "line_count": 550, "word_count_approx": 3285, "card_type": "appendix", "card_path": "cards/V03/B066.md", "in_short": "La guida di installazione rende verificabili prerequisiti e configurazione del framework.", "summary": "Setup, sdd init, variabili, MCP, adattatori, moduli, troubleshooting e riferimento CLI formano il percorso di avvio di Enterprise SDD.", "limitation": "L’installazione non è stata ripetuta in questa analisi; richiede una verifica su ambiente, release e permessi effettivi.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Productivity\", \"T:Skills & learning\", \"T:Code quality & security\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\", \"K:Dependencies and agent supply chain\", \"K:Prompt injection and malicious tools\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C06\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Prerequisites\", \"line\": 10, \"offset\": 277, \"end\": 300}, {\"level\": \"section\", \"title\": \"Environment Variables and MCP Configuration\", \"line\": 209, \"offset\": 9983, \"end\": 10036}, {\"level\": \"section\", \"title\": \"Module Installation\", \"line\": 324, \"offset\": 15036, \"end\": 15065}, {\"level\": \"section\", \"title\": \"Quick-Start Checklist\", \"line\": 525, \"offset\": 27976, \"end\": 28007}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Evaluation methodology and baselines", "K:Dependencies and agent supply chain", "K:Prompt injection and malicious tools" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/appendix_setup.tex" } }, { "id": "B067", "label": "Playbooks", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:What agents can do" ], "actual": [], "mapped": [ "C:C04", "C:C11" ], "printed": false, "summary": "I playbook trasformano la metodologia in sequenze per scenari di delivery diversi.", "limitation": "Le sei sequenze sono prescrittive; non costituiscono sei esperimenti o esiti produttivi indipendenti.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Playbooks", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/appendix_playbooks.tex", "source_sha256": "768d4170053a9348e6740381002f0a8e19fddd39149f301bf503d061127ba5ea", "line_count": 623, "word_count_approx": 3031, "card_type": "appendix", "card_path": "cards/V03/B067.md", "in_short": "I playbook trasformano la metodologia in sequenze per scenari di delivery diversi.", "summary": "Greenfield, N+1, brownfield, hotfix, batch autonomo e Three Amigos mostrano come adattare la stessa struttura al rischio e alle responsabilità.", "limitation": "Le sei sequenze sono prescrittive; non costituiscono sei esperimenti o esiti produttivi indipendenti.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:What agents can do\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Agent failure and escalation\", \"K:Long-running agent execution\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Playbook 1: Greenfield Project\", \"line\": 10, \"offset\": 264, \"end\": 304}, {\"level\": \"section\", \"title\": \"Playbook 3: Legacy Integration (Brownfield)\", \"line\": 267, \"offset\": 9956, \"end\": 10009}, {\"level\": \"section\", \"title\": \"Playbook 4: Emergency Hotfix\", \"line\": 338, \"offset\": 13527, \"end\": 13565}, {\"level\": \"section\", \"title\": \"Playbook 6: Multi-Role Team Feature (Three Amigos)\", \"line\": 549, \"offset\": 22634, \"end\": 22694}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Agent failure and escalation", "K:Long-running agent execution" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/appendix_playbooks.tex" } }, { "id": "B068", "label": "Template Reference", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C04", "C:C09", "C:C11" ], "printed": false, "summary": "I dodici template rendono esplicita la struttura documentale di Enterprise SDD.", "limitation": "Il catalogo rende visibile la densità degli artefatti; non dimostra assenza di duplicazione né sostituisce la verifica dell’intento.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Template Reference", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/appendix_templates.tex", "source_sha256": "c748b7ed2bc789816a50a82c446b253db7286e300f4f4455007fcb5313f6bbaa", "line_count": 586, "word_count_approx": 2254, "card_type": "appendix", "card_path": "cards/V03/B068.md", "in_short": "I dodici template rendono esplicita la struttura documentale di Enterprise SDD.", "summary": "Business context, spec, chiarimenti, piano, test, task, costituzione, bridge, escalation, gap, checklist e metadati forniscono una grammatica comune.", "limitation": "Il catalogo rende visibile la densità degli artefatti; non dimostra assenza di duplicazione né sostituisce la verifica dell’intento.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:Ownership and knowledge transfer\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C09\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Template 1: business-context.md\", \"line\": 10, \"offset\": 317, \"end\": 365}, {\"level\": \"section\", \"title\": \"Template 5: test-cases.md\", \"line\": 212, \"offset\": 5541, \"end\": 5583}, {\"level\": \"section\", \"title\": \"Template 8: Context Bridge\", \"line\": 366, \"offset\": 9853, \"end\": 9889}, {\"level\": \"section\", \"title\": \"Template 12: Feature Meta JSON Schema\", \"line\": 561, \"offset\": 15867, \"end\": 15914}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:Ownership and knowledge transfer", "K:Agent context and persistent memory", "K:System understanding and rationale" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/appendix_templates.tex" } }, { "id": "B069", "label": "Framework Reference Cards", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Productivity", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C03", "C:C04" ], "printed": false, "summary": "Il contenitore delle reference card collega i dodici framework ai rispettivi dettagli operativi.", "limitation": "Non espandere nuovamente i testi inclusi: questo nodo è un contenitore editoriale, non un tredicesimo framework.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Framework Reference Cards", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/appendix_reference_cards.tex", "source_sha256": "d4af5f3c964ef58efde234dcb5ca257a2595dc0cd0bd3eeeb5bd8a886b445662", "line_count": 25, "word_count_approx": 118, "card_type": "appendix", "card_path": "cards/V03/B069.md", "in_short": "Il contenitore delle reference card collega i dodici framework ai rispettivi dettagli operativi.", "summary": "Il file introduce la funzione di consultazione rapida e include dodici schede autonome, preservate individualmente nel grafo.", "limitation": "Non espandere nuovamente i testi inclusi: questo nodo è un contenitore editoriale, non un tredicesimo framework.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Productivity\", \"T:Skills & learning\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Specifications and governance\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C04\"]", "evidence_anchors_json": "[{\"level\": \"chapter\", \"title\": \"Framework Reference Cards\", \"line\": 4, \"offset\": 3, \"end\": 38}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Specifications and governance", "K:Evaluation methodology and baselines" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/appendix_reference_cards.tex" } }, { "id": "B070", "label": "Enterprise SDD", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:What agents can do" ], "actual": [], "mapped": [ "C:C04", "C:C11" ], "printed": false, "summary": "La reference card di Enterprise SDD condensa pipeline, agenti, comandi e gate.", "limitation": "È derivata dagli stessi capitoli: non una fonte indipendente né un test dei comandi elencati.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Enterprise SDD", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_enterprise_sdd.tex", "source_sha256": "25c7c152091ba72111946834e36283f3b53959f3c69c81ef74262cbeb5f9b618", "line_count": 94, "word_count_approx": 929, "card_type": "reference_card", "card_path": "cards/V03/B070.md", "in_short": "La reference card di Enterprise SDD condensa pipeline, agenti, comandi e gate.", "summary": "La scheda fornisce un promemoria operativo della proposta sviluppata nei capitoli dedicati.", "limitation": "È derivata dagli stessi capitoli: non una fonte indipendente né un test dei comandi elencati.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:What agents can do\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Ownership and knowledge transfer\", \"K:Agent failure and escalation\", \"K:Long-running agent execution\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Enterprise SDD\", \"line\": 4, \"offset\": 3, \"end\": 27}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Ownership and knowledge transfer", "K:Agent failure and escalation", "K:Long-running agent execution" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_enterprise_sdd.tex" } }, { "id": "B071", "label": "Spec Kit", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C04" ], "printed": false, "summary": "La reference card di Spec Kit collega comandi, artefatti e sequenza di lavoro.", "limitation": "Il riepilogo è datato dalla versione del testo e non equivale a verifica di compatibilità dell’installazione corrente.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Spec Kit", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_speckit.tex", "source_sha256": "dce70bac1e140c115657f1304e602bce11c07bbff82419103075243d0f1726e9", "line_count": 89, "word_count_approx": 972, "card_type": "reference_card", "card_path": "cards/V03/B071.md", "in_short": "La reference card di Spec Kit collega comandi, artefatti e sequenza di lavoro.", "summary": "La scheda riassume il ciclo della toolchain e il ruolo dei file prodotti come supporto alla consultazione.", "limitation": "Il riepilogo è datato dalla versione del testo e non equivale a verifica di compatibilità dell’installazione corrente.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:System understanding and rationale\"]", "editorial_chapters_json": "[\"C:C04\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Spec Kit\", \"line\": 4, \"offset\": 3, \"end\": 21}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:System understanding and rationale" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_speckit.tex" } }, { "id": "B072", "label": "OpenSpec (openspec.dev)", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C04", "C:C05" ], "printed": false, "summary": "La reference card di OpenSpec rende leggibile il cambiamento tramite pacchetti e delta.", "limitation": "Le azioni e gli effetti di lifecycle vanno mantenuti distinti; la scheda non autorizza a inferire un processo rigidamente sequenziale.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "OpenSpec (openspec.dev)", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_openspec.tex", "source_sha256": "1b144458ca9d3d08ad0d48181ef535eb484b8a1a09a138b8c1a52068f58c961d", "line_count": 143, "word_count_approx": 1389, "card_type": "reference_card", "card_path": "cards/V03/B072.md", "in_short": "La reference card di OpenSpec rende leggibile il cambiamento tramite pacchetti e delta.", "summary": "La scheda descrive comandi di assistenza, artefatti del change package e integrazione nel repository brownfield.", "limitation": "Le azioni e gli effetti di lifecycle vanno mantenuti distinti; la scheda non autorizza a inferire un processo rigidamente sequenziale.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:Code quality & security\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Architecture and modularity\", \"K:Technical debt and complexity\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C05\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"OpenSpec (openspec.dev)\", \"line\": 4, \"offset\": 3, \"end\": 36}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Architecture and modularity", "K:Technical debt and complexity" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_openspec.tex" } }, { "id": "B073", "label": "BMAD Method", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure", "T:Verification & review" ], "actual": [], "mapped": [ "C:C04", "C:C11" ], "printed": false, "summary": "La reference card di BMAD evidenzia agenti persona e artefatti per fase.", "limitation": "Personae differenti non dimostrano fonti di evidenza indipendenti o controlli realmente separati.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "BMAD Method", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_bmad.tex", "source_sha256": "f8ad28cb7ef76ec9f7d2ca975324fc8affc7639b9e75678edb6ec2c6043e5c8c", "line_count": 85, "word_count_approx": 736, "card_type": "reference_card", "card_path": "cards/V03/B073.md", "in_short": "La reference card di BMAD evidenzia agenti persona e artefatti per fase.", "summary": "La scheda condensa il percorso Analyst–PM–Architect–Developer e i relativi output.", "limitation": "Personae differenti non dimostrano fonti di evidenza indipendenti o controlli realmente separati.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Engineering structure\", \"T:Verification & review\"]", "topics_json": "[\"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\", \"K:Specifications and governance\", \"K:Independent verification and oracles\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"BMAD Method\", \"line\": 4, \"offset\": 3, \"end\": 24}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution", "K:Specifications and governance", "K:Independent verification and oracles" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_bmad.tex" } }, { "id": "B074", "label": "AI-RPI Protocol", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:What agents can do", "T:Verification & review" ], "actual": [], "mapped": [ "C:C04", "C:C11" ], "printed": false, "summary": "La reference card di AI-RPI esplicita ricerca, piano e implementazione adattiva.", "limitation": "La governance adattiva proposta va verificata sui criteri che attivano o riducono i controlli.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "AI-RPI Protocol", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_airpi.tex", "source_sha256": "b662d052eee3620bd9375923ce1694e16c5ebbf77692cc9a3f12c0027744425f", "line_count": 82, "word_count_approx": 782, "card_type": "reference_card", "card_path": "cards/V03/B074.md", "in_short": "La reference card di AI-RPI esplicita ricerca, piano e implementazione adattiva.", "summary": "Livelli di protocollo e artefatti permettono di applicare rigore diverso senza perdere il filo del lavoro.", "limitation": "La governance adattiva proposta va verificata sui criteri che attivano o riducono i controlli.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:What agents can do\", \"T:Verification & review\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"AI-RPI Protocol\", \"line\": 4, \"offset\": 3, \"end\": 28}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_airpi.tex" } }, { "id": "B075", "label": "ralph-copilot", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Engineering structure", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C03", "C:C09" ], "printed": false, "summary": "La reference card di ralph-copilot concentra il loop in ruoli e memoria su file.", "limitation": "Il ruolo Reviewer non garantisce indipendenza dalla generazione; verificare gli input e i poteri reali del controllo.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "ralph-copilot", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_ralph.tex", "source_sha256": "3deae3fe3e14915cd1b324bcc1c711f46488c4155061dac55cbd34113113f697", "line_count": 77, "word_count_approx": 703, "card_type": "reference_card", "card_path": "cards/V03/B075.md", "in_short": "La reference card di ralph-copilot concentra il loop in ruoli e memoria su file.", "summary": "Planner, Coordinator, Executor e Reviewer vengono messi in relazione con stato persistente e ripetizione del ciclo.", "limitation": "Il ruolo Reviewer non garantisce indipendenza dalla generazione; verificare gli input e i poteri reali del controllo.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Engineering structure\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"ralph-copilot\", \"line\": 4, \"offset\": 3, \"end\": 26}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles", "K:Agent context and persistent memory", "K:System understanding and rationale" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_ralph.tex" } }, { "id": "B076", "label": "Spec Loop Engine", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Engineering structure", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C03", "C:C07" ], "printed": false, "summary": "La reference card di Spec Loop Engine separa executor e verifier su uno stato durevole.", "limitation": "La separazione dei componenti è architetturale; resta da dimostrare l’indipendenza del criterio usato per accettare l’esito.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Spec Loop Engine", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_specloop.tex", "source_sha256": "6b98bfe86bc7e9ee7382830193b3ae042c358f095e9dab38042aca94b00234b6", "line_count": 85, "word_count_approx": 801, "card_type": "reference_card", "card_path": "cards/V03/B076.md", "in_short": "La reference card di Spec Loop Engine separa executor e verifier su uno stato durevole.", "summary": "Il modello YAML/JSON e il ciclo a due passi rendono espliciti specifica, esecuzione e controllo.", "limitation": "La separazione dei componenti è architetturale; resta da dimostrare l’indipendenza del criterio usato per accettare l’esito.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Engineering structure\", \"T:Code quality & security\"]", "topics_json": "[\"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C07\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Spec Loop Engine\", \"line\": 4, \"offset\": 3, \"end\": 29}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles", "K:Evaluation methodology and baselines" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_specloop.tex" } }, { "id": "B077", "label": "GSD v1", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:System knowledge & memory", "T:What agents can do" ], "actual": [], "mapped": [ "C:C04" ], "printed": false, "summary": "La reference card di GSD v1 mostra il percorso leggero Discuss–Plan–Execute–Verify.", "limitation": "La brevità del protocollo non misura overhead reale o qualità della verifica.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "GSD v1", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_gsd_v1.tex", "source_sha256": "c95efa6fc3330eaddd5f9340d9678768d44cebebd8053a935539f6c4ce6d267e", "line_count": 84, "word_count_approx": 730, "card_type": "reference_card", "card_path": "cards/V03/B077.md", "in_short": "La reference card di GSD v1 mostra il percorso leggero Discuss–Plan–Execute–Verify.", "summary": "Comandi e artefatti sono presentati come protocollo minimo per disciplinare il lavoro.", "limitation": "La brevità del protocollo non misura overhead reale o qualità della verifica.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:System knowledge & memory\", \"T:What agents can do\"]", "topics_json": "[\"K:Specifications and governance\", \"K:System understanding and rationale\", \"K:Developer workflow and adoption\"]", "editorial_chapters_json": "[\"C:C04\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"GSD v1\", \"line\": 4, \"offset\": 3, \"end\": 19}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:System understanding and rationale", "K:Developer workflow and adoption" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_gsd_v1.tex" } }, { "id": "B078", "label": "GSD-2", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C03", "C:C04", "C:C09" ], "printed": false, "summary": "La reference card di GSD-2 mette in primo piano stato persistente e modalità automatica.", "limitation": "Auto-mode non implica autorizzazione illimitata; arresti, resume e confini vanno verificati nella versione concreta.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "GSD-2", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_gsd2.tex", "source_sha256": "9ae80348026eb73434b6bc9491d1dcf8a1dbf39a6d91d0c1ea908d4ea970b4c5", "line_count": 88, "word_count_approx": 944, "card_type": "reference_card", "card_path": "cards/V03/B078.md", "in_short": "La reference card di GSD-2 mette in primo piano stato persistente e modalità automatica.", "summary": "Il file descrive sottocomandi, artefatti su disco e transizioni della macchina a stati.", "limitation": "Auto-mode non implica autorizzazione illimitata; arresti, resume e confini vanno verificati nella versione concreta.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Engineering structure\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Agent failure and escalation\", \"K:Long-running agent execution\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C04\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"GSD-2\", \"line\": 4, \"offset\": 3, \"end\": 18}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent failure and escalation", "K:Long-running agent execution", "K:Agent context and persistent memory", "K:System understanding and rationale" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_gsd2.tex" } }, { "id": "B079", "label": "Karpathy-Inspired Skills", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Skills & learning", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C04", "C:C10" ], "printed": false, "summary": "Le skill ispirate a Karpathy sono principi comportamentali, non una pipeline SDLC.", "limitation": "Non confrontare questo insieme di principi con una toolchain come se avesse la stessa unità funzionale o le stesse garanzie.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Karpathy-Inspired Skills", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_karpathy.tex", "source_sha256": "3c84a699d6381f3230a19d991e5b763b7f0a681bad6c2b161040ea230f8dcd3c", "line_count": 81, "word_count_approx": 599, "card_type": "reference_card", "card_path": "cards/V03/B079.md", "in_short": "Le skill ispirate a Karpathy sono principi comportamentali, non una pipeline SDLC.", "summary": "La scheda separa disciplina di lavoro, comportamento dell’agente e assenza di comandi o artefatti di processo propri.", "limitation": "Non confrontare questo insieme di principi con una toolchain come se avesse la stessa unità funzionale o le stesse garanzie.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Skills & learning\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Specifications and governance\", \"K:Skills formation and teaching\", \"K:Cognitive offloading and skill debt\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C10\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Karpathy-Inspired Skills\", \"line\": 4, \"offset\": 3, \"end\": 37}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Specifications and governance", "K:Skills formation and teaching", "K:Cognitive offloading and skill debt" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_karpathy.tex" } }, { "id": "B080", "label": "Agent Skills", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:System knowledge & memory", "T:What agents can do", "T:Engineering structure" ], "actual": [], "mapped": [ "C:C03", "C:C09" ], "printed": false, "summary": "Agent Skills usa composizione e caricamento progressivo per fornire capacità contestuali.", "limitation": "Una skill può essere obsoleta o non attendibile; presenza nel catalogo e caricamento corretto non equivalgono a validazione.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Agent Skills", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_agent_skills.tex", "source_sha256": "397093258847e4d6d6e27af6d0e1c413074401785e7ff2c5f83a05d636d62790", "line_count": 86, "word_count_approx": 813, "card_type": "reference_card", "card_path": "cards/V03/B080.md", "in_short": "Agent Skills usa composizione e caricamento progressivo per fornire capacità contestuali.", "summary": "La scheda organizza comandi, personas e struttura SKILL.md come componenti riutilizzabili.", "limitation": "Una skill può essere obsoleta o non attendibile; presenza nel catalogo e caricamento corretto non equivalgono a validazione.", "primary_theme": "T:System knowledge & memory", "streams_json": "[\"T:System knowledge & memory\", \"T:What agents can do\", \"T:Engineering structure\"]", "topics_json": "[\"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Dependencies and agent supply chain\", \"K:Developer workflow and adoption\", \"K:Specifications and governance\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Agent Skills\", \"line\": 4, \"offset\": 3, \"end\": 25}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Dependencies and agent supply chain", "K:Developer workflow and adoption", "K:Specifications and governance" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_agent_skills.tex" } }, { "id": "B081", "label": "AIDD", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Code quality & security", "T:Engineering structure", "T:Verification & review" ], "actual": [], "mapped": [ "C:C04", "C:C05" ], "printed": false, "summary": "AIDD organizza il lavoro a partire da visione e decisioni architetturali.", "limitation": "Il profilo si riferisce alla versione indicata nel testo; non è un benchmark né una verifica degli strumenti attuali.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "AIDD", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_aidd.tex", "source_sha256": "8dfbbd6274c5de05c2886c5945740e7e4d5572836ef1acbdf9e6d45b8202b6ed", "line_count": 89, "word_count_approx": 825, "card_type": "reference_card", "card_path": "cards/V03/B081.md", "in_short": "AIDD organizza il lavoro a partire da visione e decisioni architetturali.", "summary": "La scheda sintetizza comandi, skill SudoLang e orchestrazione della pipeline backend descritta nel libro.", "limitation": "Il profilo si riferisce alla versione indicata nel testo; non è un benchmark né una verifica degli strumenti attuali.", "primary_theme": "T:Code quality & security", "streams_json": "[\"T:Code quality & security\", \"T:Engineering structure\", \"T:Verification & review\"]", "topics_json": "[\"K:Architecture and modularity\", \"K:Technical debt and complexity\", \"K:Specifications and governance\", \"K:Independent verification and oracles\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C05\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"AIDD\", \"line\": 4, \"offset\": 3, \"end\": 17}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Architecture and modularity", "K:Technical debt and complexity", "K:Specifications and governance", "K:Independent verification and oracles" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/reference_cards/appendix_aidd.tex" } }, { "id": "B082", "label": "Previous Books and Author", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do" ], "actual": [], "mapped": [ "C:C01" ], "printed": false, "summary": "L’apparato finale documenta serie, attribuzioni ed opere richiamate.", "limitation": "La presenza nell’apparato bibliografico non equivale a citazione in ogni capitolo, né a conferma delle tesi del framework.", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Previous Books and Author", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V03", "volume_title": "Non-Deterministic Spec-driven Development — 2nd edition", "source_path": "AI_SDD_ENG_V2/appendices/appendix_references.tex", "source_sha256": "28ffc18780ee3435dc7b39370cad24185fa5004f911e42596042109b9b5bbb97", "line_count": 116, "word_count_approx": 2352, "card_type": "bibliographic_apparatus", "card_path": "cards/V03/B082.md", "in_short": "L’apparato finale documenta serie, attribuzioni ed opere richiamate.", "summary": "Il file collega i volumi dell’autore e raccoglie fonti degli epigrafi e lavori citati, rendendo tracciabile la genealogia editoriale.", "limitation": "La presenza nell’apparato bibliografico non equivale a citazione in ogni capitolo, né a conferma delle tesi del framework.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Human-agent collaboration\"]", "editorial_chapters_json": "[\"C:C01\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Books in the Series\", \"line\": 7, \"offset\": 64, \"end\": 94}, {\"level\": \"section\", \"title\": \"Sources for Chapter Epigraphs\", \"line\": 66, \"offset\": 3858, \"end\": 3898}, {\"level\": \"section\", \"title\": \"Cited Works\", \"line\": 115, \"offset\": 18281, \"end\": 18303}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Human-agent collaboration" ], "volumes": [ "V03" ], "book": { "volume": "V03", "title": "Non-Deterministic Spec-Driven Development", "subtitle": "Enterprise Edition -- From Public Frameworks to Enterprise Practice", "url": "https://www.amazon.it/dp/B0GZZDFMYF", "edition": "2", "source_path": "AI_SDD_ENG_V2/appendices/appendix_references.tex" } }, { "id": "B083", "label": "Introduzioni e postfazione — The Autonomous Future", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure", "T:Jobs & roles", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C01", "C:C03", "C:C11" ], "printed": false, "summary": "Le introduzioni alle parti e la postfazione presentano il passaggio dall’assistenza interattiva alla delega continuativa.", "limitation": "Il futuro autonomo è una prospettiva editoriale, non un risultato empirico; i figli inclusi restano unità separate.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Introduzioni e postfazione — The Autonomous Future", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/main.tex", "source_sha256": "15e7b52d24d8b4a9b0fd959d40227c60942ae4aa0fd071fbe674e0a888d5cad2", "line_count": 413, "word_count_approx": 2112, "card_type": "structural_synthesis", "card_path": "cards/V04/B083.md", "in_short": "Le introduzioni alle parti e la postfazione presentano il passaggio dall’assistenza interattiva alla delega continuativa.", "summary": "Il main contiene raccordi didattici e una conclusione sull’evoluzione del lavoro, distinti dagli input dei capitoli; le porzioni biografiche non alimentano relazioni tematiche.", "limitation": "Il futuro autonomo è una prospettiva editoriale, non un risultato empirico; i figli inclusi restano unità separate.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Engineering structure\", \"T:Jobs & roles\", \"T:Skills & learning\"]", "topics_json": "[\"K:Agent failure and escalation\", \"K:Long-running agent execution\", \"K:Employment and early-career roles\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C03\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"chapter\", \"title\": \"Afterword: The Autonomous Future\", \"line\": 204, \"offset\": 10521, \"end\": 10564}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent failure and escalation", "K:Long-running agent execution", "K:Employment and early-career roles", "K:Ownership and knowledge transfer" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/main.tex" } }, { "id": "B084", "label": "Preface", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure", "T:Productivity", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C01", "C:C03", "C:C11" ], "printed": false, "summary": "La prefazione definisce destinatari, percorso di lettura e precauzioni per costruire agenti autonomi.", "limitation": "L’attribuzione complessiva non prova che ogni esempio derivi da una fonte specifica né che l’automazione sia sicura in produzione.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Preface", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/frontmatter.tex", "source_sha256": "3af22b719d0303a2013767e4c5e25724708b593c3a144393dc78f940946cf1ab", "line_count": 325, "word_count_approx": 3034, "card_type": "reading_guide", "card_path": "cards/V04/B084.md", "in_short": "La prefazione definisce destinatari, percorso di lettura e precauzioni per costruire agenti autonomi.", "summary": "Il volume si presenta come seguito dello SDD e attribuisce contributi a Huntley, Yegge, Carson e Osmani; documenta il contesto practitioner della sintesi.", "limitation": "L’attribuzione complessiva non prova che ogni esempio derivi da una fonte specifica né che l’automazione sia sicura in produzione.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Engineering structure\", \"T:Productivity\", \"T:Skills & learning\"]", "topics_json": "[\"K:Agent failure and escalation\", \"K:Long-running agent execution\", \"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C03\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"chapter\", \"title\": \"Preface\", \"line\": 78, \"offset\": 3574, \"end\": 3592}, {\"level\": \"subsection\", \"title\": \"Who This Book Is For\", \"line\": 146, \"offset\": 8457, \"end\": 8491}, {\"level\": \"subsubsection\", \"title\": \"The Recommended Path\", \"line\": 191, \"offset\": 10969, \"end\": 11006}, {\"level\": \"subsection\", \"title\": \"The Experts Who Shaped This Book\", \"line\": 292, \"offset\": 17629, \"end\": 17675}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent failure and escalation", "K:Long-running agent execution", "K:Developer workflow and adoption", "K:Evaluation methodology and baselines" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/frontmatter.tex" } }, { "id": "B085", "label": "The Continuous Development Revolution", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure", "T:Verification & review", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C01", "C:C03", "C:C08" ], "printed": false, "summary": "Lo sviluppo continuativo combina esecuzione autonoma, contesto rinnovato e una frontiera umana di approvazione.", "limitation": "La narrazione del lavoro da 50.000 dollari svolto con 297 dollari di API non misura ROI comparabile: mancano costo umano, baseline e denominatore della qualità accettata.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Continuous Development Revolution", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter01_revolution.tex", "source_sha256": "59c19cee9311375de9d39144d6318c63f37871af01d55d8d8a1774aa99d5304f", "line_count": 398, "word_count_approx": 3879, "card_type": "chapter", "card_path": "cards/V04/B085.md", "in_short": "Lo sviluppo continuativo combina esecuzione autonoma, contesto rinnovato e una frontiera umana di approvazione.", "summary": "Il capitolo lega Ralph, apprendimento cumulativo, specifiche, infrastruttura di test e controllo di versione a un flusso che prepara lavoro per la revisione.", "limitation": "La narrazione del lavoro da 50.000 dollari svolto con 297 dollari di API non misura ROI comparabile: mancano costo umano, baseline e denominatore della qualità accettata.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Engineering structure\", \"T:Verification & review\", \"T:Code quality & security\"]", "topics_json": "[\"K:Agent failure and escalation\", \"K:Long-running agent execution\", \"K:Independent verification and oracles\", \"K:Review capacity and maintainers\", \"K:AI as reviewer\", \"K:Automation bias and trust\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C03\", \"C:C08\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"From Reactive to Proactive AI\", \"line\": 25, \"offset\": 1824, \"end\": 1863}, {\"level\": \"section\", \"title\": \"The Continuous Loop Concept\", \"line\": 117, \"offset\": 10440, \"end\": 10477}, {\"level\": \"section\", \"title\": \"Prerequisites for Success\", \"line\": 246, \"offset\": 18642, \"end\": 18677}, {\"level\": \"section\", \"title\": \"What This Book Will Teach You\", \"line\": 319, \"offset\": 25530, \"end\": 25569}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent failure and escalation", "K:Long-running agent execution", "K:Independent verification and oracles", "K:Review capacity and maintainers", "K:AI as reviewer", "K:Automation bias and trust" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter01_revolution.tex" } }, { "id": "B086", "label": "The Philosophy of Deterministic Failure", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Code quality & security", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C03", "C:C07", "C:C09" ], "printed": false, "summary": "Progettare per il fallimento prevedibile significa fornire segnali di rifiuto e stato recuperabile.", "limitation": "Il controllo respinge soltanto ciò che sa riconoscere; LLM-as-judge e soglia contestuale 40–60% richiedono qualificazione e verifica nel contesto d’uso.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Philosophy of Deterministic Failure", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter02_philosophy.tex", "source_sha256": "e5877964bca2d35f772986e072b4087b75792a15e89705c2015c204ac06d6dfe", "line_count": 490, "word_count_approx": 5021, "card_type": "chapter", "card_path": "cards/V04/B086.md", "in_short": "Progettare per il fallimento prevedibile significa fornire segnali di rifiuto e stato recuperabile.", "summary": "Backpressure, test, contesto fresco, piano modificabile e memoria su disco spostano l’attenzione dal prompt perfetto alle condizioni che bloccano lavoro invalido.", "limitation": "Il controllo respinge soltanto ciò che sa riconoscere; LLM-as-judge e soglia contestuale 40–60% richiedono qualificazione e verifica nel contesto d’uso.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Code quality & security\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Agent failure and escalation\", \"K:Evaluation methodology and baselines\", \"K:Independent verification and oracles\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C07\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Embracing Predictable Failure\", \"line\": 18, \"offset\": 1425, \"end\": 1464}, {\"level\": \"section\", \"title\": \"The Backpressure Model\", \"line\": 96, \"offset\": 7874, \"end\": 7906}, {\"level\": \"section\", \"title\": \"Core Principles for Autonomous Systems\", \"line\": 289, \"offset\": 24588, \"end\": 24636}, {\"level\": \"section\", \"title\": \"Common Anti-Patterns to Avoid\", \"line\": 444, \"offset\": 34444, \"end\": 34483}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent failure and escalation", "K:Evaluation methodology and baselines", "K:Independent verification and oracles", "K:Agent context and persistent memory", "K:System understanding and rationale" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter02_philosophy.tex" } }, { "id": "B087", "label": "The Developer Evolution Model", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure", "T:Skills & learning", "T:System knowledge & memory", "T:Jobs & roles" ], "actual": [], "mapped": [ "C:C03", "C:C10", "C:C11" ], "printed": false, "summary": "Il modello a nove stadi rende discutibili competenze e prerequisiti della delega agli agenti.", "limitation": "È una tassonomia didattica, non una scala validata. Il confronto con gli otto stadi attribuiti a Yegge nel corpus P richiede distinguere adattamento e citazione letterale.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Developer Evolution Model", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter03_evolution.tex", "source_sha256": "8100340c3982a08db0283347394523f0ebfca3a026102da8fce8992f9d085db4", "line_count": 395, "word_count_approx": 4034, "card_type": "chapter", "card_path": "cards/V04/B087.md", "in_short": "Il modello a nove stadi rende discutibili competenze e prerequisiti della delega agli agenti.", "summary": "La progressione è usata per autovalutare pratiche di sviluppo, trasformazione delle abilità e preparazione a sistemi autonomi.", "limitation": "È una tassonomia didattica, non una scala validata. Il confronto con gli otto stadi attribuiti a Yegge nel corpus P richiede distinguere adattamento e citazione letterale.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Engineering structure\", \"T:Skills & learning\", \"T:System knowledge & memory\", \"T:Jobs & roles\"]", "topics_json": "[\"K:Agent failure and escalation\", \"K:Long-running agent execution\", \"K:Skills formation and teaching\", \"K:Cognitive offloading and skill debt\", \"K:Employment and early-career roles\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C10\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Nine Stages of AI-Assisted Development\", \"line\": 17, \"offset\": 1584, \"end\": 1636}, {\"level\": \"section\", \"title\": \"Assessing Your Current Stage\", \"line\": 182, \"offset\": 15152, \"end\": 15190}, {\"level\": \"section\", \"title\": \"The Skills Transformation\", \"line\": 264, \"offset\": 21273, \"end\": 21308}, {\"level\": \"section\", \"title\": \"Prerequisites for This Book\", \"line\": 321, \"offset\": 25248, \"end\": 25285}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent failure and escalation", "K:Long-running agent execution", "K:Skills formation and teaching", "K:Cognitive offloading and skill debt", "K:Employment and early-career roles", "K:Ownership and knowledge transfer" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter03_evolution.tex" } }, { "id": "B088", "label": "The Ralph Loop Pattern", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Engineering structure", "T:System knowledge & memory", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C03", "C:C07", "C:C09" ], "printed": false, "summary": "Il pattern Ralph ripete un incarico con contesto fresco e progresso conservato fuori dal modello.", "limitation": "Un segnale di completamento o un test passato non garantiscono che l’intento sia soddisfatto; i comandi non sono stati eseguiti in questa analisi.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Ralph Loop Pattern", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter04_ralph.tex", "source_sha256": "bc33f212af1d60604b0a868e40a9cd2c1d385092dbc68579a76bebfc1ebb2d03", "line_count": 703, "word_count_approx": 4095, "card_type": "chapter", "card_path": "cards/V04/B088.md", "in_short": "Il pattern Ralph ripete un incarico con contesto fresco e progresso conservato fuori dal modello.", "summary": "Anatomia, ciclo di iterazione, varianti implementative, primo esercizio e diagnosi iniziale trasformano il loop in una procedura concreta.", "limitation": "Un segnale di completamento o un test passato non garantiscono che l’intento sia soddisfatto; i comandi non sono stati eseguiti in questa analisi.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Engineering structure\", \"T:System knowledge & memory\", \"T:Code quality & security\"]", "topics_json": "[\"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C07\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Anatomy of a Ralph Loop\", \"line\": 20, \"offset\": 1582, \"end\": 1615}, {\"level\": \"section\", \"title\": \"Implementation Options\", \"line\": 328, \"offset\": 16885, \"end\": 16917}, {\"level\": \"section\", \"title\": \"Your First Ralph Loop\", \"line\": 416, \"offset\": 21659, \"end\": 21690}, {\"level\": \"section\", \"title\": \"Troubleshooting Common Issues\", \"line\": 650, \"offset\": 30115, \"end\": 30154}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles", "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Evaluation methodology and baselines" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter04_ralph.tex" } }, { "id": "B089", "label": "Task Decomposition for Machines", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:System knowledge & memory", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C04", "C:C07" ], "printed": false, "summary": "La decomposizione rende una richiesta eseguibile attraverso unità limitate e criteri verificabili.", "limitation": "La piccola dimensione dei task non risolve dipendenze e ambiguità; la specifica deve restare indipendente dall’implementazione candidata.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Task Decomposition for Machines", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter05_decomposition.tex", "source_sha256": "4891c821bb6f814c065b12abb79194075ef3e9d7536433cbeff318c1dbf87b7f", "line_count": 703, "word_count_approx": 4538, "card_type": "chapter", "card_path": "cards/V04/B089.md", "in_short": "La decomposizione rende una richiesta eseguibile attraverso unità limitate e criteri verificabili.", "summary": "Dal PRD al piano strutturato, il capitolo descrive caratteristiche dei task, schema del file e criteri di accettazione con un esempio completo.", "limitation": "La piccola dimensione dei task non risolve dipendenze e ambiguità; la specifica deve restare indipendente dall’implementazione candidata.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:System knowledge & memory\", \"T:Code quality & security\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:System understanding and rationale\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C07\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"From PRD to Executable Tasks\", \"line\": 24, \"offset\": 2132, \"end\": 2170}, {\"level\": \"section\", \"title\": \"The Plan File Schema\", \"line\": 225, \"offset\": 16456, \"end\": 16486}, {\"level\": \"section\", \"title\": \"Writing Effective Acceptance Criteria\", \"line\": 352, \"offset\": 21732, \"end\": 21779}, {\"level\": \"section\", \"title\": \"A Complete Worked Example\", \"line\": 506, \"offset\": 29459, \"end\": 29494}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:System understanding and rationale", "K:Evaluation methodology and baselines" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter05_decomposition.tex" } }, { "id": "B090", "label": "Memory and Context Engineering", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:System knowledge & memory", "T:What agents can do", "T:Engineering structure", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C09", "C:C10" ], "printed": false, "summary": "Quattro canali di memoria consentono al loop di conservare conoscenza senza mantenere una sessione infinita.", "limitation": "Accumulo non significa apprendimento corretto: servono provenienza, rimozione di lezioni errate e invalidazione quando cambia l’intenzione.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Memory and Context Engineering", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter06_memory.tex", "source_sha256": "46beb0d7a4f5a99ab0393394551dc7016d76c6d09e941ea203076ef24e723c2a", "line_count": 591, "word_count_approx": 5678, "card_type": "chapter", "card_path": "cards/V04/B090.md", "in_short": "Quattro canali di memoria consentono al loop di conservare conoscenza senza mantenere una sessione infinita.", "summary": "AGENTS.md, apprendimento cumulativo e strategie di iniezione del contesto organizzano stato operativo e conoscenza durevole.", "limitation": "Accumulo non significa apprendimento corretto: servono provenienza, rimozione di lezioni errate e invalidazione quando cambia l’intenzione.", "primary_theme": "T:System knowledge & memory", "streams_json": "[\"T:System knowledge & memory\", \"T:What agents can do\", \"T:Engineering structure\", \"T:Skills & learning\"]", "topics_json": "[\"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Skills formation and teaching\", \"K:Cognitive offloading and skill debt\"]", "editorial_chapters_json": "[\"C:C09\", \"C:C10\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Four Channels of Memory\", \"line\": 18, \"offset\": 1557, \"end\": 1594}, {\"level\": \"section\", \"title\": \"Building AGENTS.md\", \"line\": 136, \"offset\": 10099, \"end\": 10127}, {\"level\": \"section\", \"title\": \"Context Injection Strategies\", \"line\": 420, \"offset\": 30296, \"end\": 30334}, {\"level\": \"section\", \"title\": \"Memory Injection Technique\", \"line\": 510, \"offset\": 38493, \"end\": 38529}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Skills formation and teaching", "K:Cognitive offloading and skill debt" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter06_memory.tex" } }, { "id": "B091", "label": "Quality Gates and Backpressure", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Verification & review", "T:Code quality & security", "T:What agents can do" ], "actual": [], "mapped": [ "C:C07", "C:C08" ], "printed": false, "summary": "Quality gate e backpressure trasformano gli errori in rifiuti operativi anziché in accettazioni apparenti.", "limitation": "La separazione dei ruoli e l’auto-critica non garantiscono indipendenza; un caso illustrativo non misura la probabilità residua di difetto.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Quality Gates and Backpressure", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter07_quality.tex", "source_sha256": "9eff56f237ad4042054343dcb9d2fb82b5b74ac0a6c30d2b80e683c84abb7445", "line_count": 661, "word_count_approx": 4311, "card_type": "chapter", "card_path": "cards/V04/B091.md", "in_short": "Quality gate e backpressure trasformano gli errori in rifiuti operativi anziché in accettazioni apparenti.", "summary": "Il capitolo compone test, verifica, gestione degli stalli e giudici linguistici, con configurazioni ed esempio applicativo.", "limitation": "La separazione dei ruoli e l’auto-critica non garantiscono indipendenza; un caso illustrativo non misura la probabilità residua di difetto.", "primary_theme": "T:Verification & review", "streams_json": "[\"T:Verification & review\", \"T:Code quality & security\", \"T:What agents can do\"]", "topics_json": "[\"K:Independent verification and oracles\", \"K:Evaluation methodology and baselines\", \"K:AI as reviewer\", \"K:Automation bias and trust\", \"K:Agent failure and escalation\"]", "editorial_chapters_json": "[\"C:C07\", \"C:C08\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Quality Checks in the Loop\", \"line\": 18, \"offset\": 1498, \"end\": 1534}, {\"level\": \"section\", \"title\": \"The Verification Loop Pattern\", \"line\": 241, \"offset\": 15011, \"end\": 15050}, {\"level\": \"section\", \"title\": \"LLM-as-Judge for Subjective Criteria\", \"line\": 399, \"offset\": 24035, \"end\": 24081}, {\"level\": \"section\", \"title\": \"Case Study: Quality Gates in Action\", \"line\": 599, \"offset\": 32676, \"end\": 32721}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Independent verification and oracles", "K:Evaluation methodology and baselines", "K:AI as reviewer", "K:Automation bias and trust", "K:Agent failure and escalation" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter07_quality.tex" } }, { "id": "B092", "label": "Advanced Prompting for Autonomous Systems", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:What agents can do", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C04", "C:C07", "C:C09" ], "printed": false, "summary": "I prompt autonomi combinano chiarimento preventivo, vincoli stratificati e prospettive di revisione.", "limitation": "Il punto critico è la correlazione dell’errore: chiedere al medesimo sistema di criticarsi può migliorare un candidato ma non costituisce prova indipendente.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Advanced Prompting for Autonomous Systems", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter08_prompting.tex", "source_sha256": "0449f462ab298c6f8c9d127bc0fc0f20f5d3aa027e78315d6a2fff9be98e4bda", "line_count": 735, "word_count_approx": 4648, "card_type": "chapter", "card_path": "cards/V04/B092.md", "in_short": "I prompt autonomi combinano chiarimento preventivo, vincoli stratificati e prospettive di revisione.", "summary": "Reverse prompting, constraint cascades, role stacking e self-critique diventano tecniche selezionabili in base al problema del task.", "limitation": "Il punto critico è la correlazione dell’errore: chiedere al medesimo sistema di criticarsi può migliorare un candidato ma non costituisce prova indipendente.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:What agents can do\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Independent verification and oracles\", \"K:AI as reviewer\", \"K:Automation bias and trust\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Dependencies and agent supply chain\"]", "editorial_chapters_json": "[\"C:C04\", \"C:C07\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Reverse Prompting: Clarification Before Execution\", \"line\": 26, \"offset\": 2248, \"end\": 2307}, {\"level\": \"section\", \"title\": \"Role Stacking: Multiple Perspectives\", \"line\": 284, \"offset\": 18183, \"end\": 18229}, {\"level\": \"section\", \"title\": \"The Verification Loop: Self-Critique Patterns\", \"line\": 393, \"offset\": 23908, \"end\": 23963}, {\"level\": \"section\", \"title\": \"Putting It Together: Technique Selection\", \"line\": 672, \"offset\": 35624, \"end\": 35674}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Independent verification and oracles", "K:AI as reviewer", "K:Automation bias and trust", "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Dependencies and agent supply chain" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter08_prompting.tex" } }, { "id": "B093", "label": "Agentic Patterns in Practice", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Verification & review", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C03", "C:C06", "C:C11" ], "printed": false, "summary": "Il catalogo organizza pattern agentici per controllo, memoria, feedback, affidabilità e sicurezza.", "limitation": "Una tassonomia ampia non dimostra il beneficio di aggiungere componenti; coordinamento e costi devono giustificare ogni meccanismo.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Agentic Patterns in Practice", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter09_patterns.tex", "source_sha256": "3c88f9e5d7b0288e46de69503668ac80ecb7778165d80bf74eab10da30c4e2ba", "line_count": 558, "word_count_approx": 4607, "card_type": "chapter", "card_path": "cards/V04/B093.md", "in_short": "Il catalogo organizza pattern agentici per controllo, memoria, feedback, affidabilità e sicurezza.", "summary": "Il capitolo propone composizione e selezione dei pattern, esplicitando anche collaborazione, ambiente e aggiornamento del repertorio.", "limitation": "Una tassonomia ampia non dimostra il beneficio di aggiungere componenti; coordinamento e costi devono giustificare ogni meccanismo.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Verification & review\", \"T:Code quality & security\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Specifications and governance\", \"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\", \"K:Prompt injection and malicious tools\", \"K:Vulnerability and secure code\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C06\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Pattern Landscape\", \"line\": 20, \"offset\": 1954, \"end\": 1985}, {\"level\": \"section\", \"title\": \"Learning \\\\& Adaptation Patterns\", \"line\": 252, \"offset\": 17570, \"end\": 17611}, {\"level\": \"section\", \"title\": \"Tool Use \\\\& Environment Patterns\", \"line\": 327, \"offset\": 23849, \"end\": 23891}, {\"level\": \"section\", \"title\": \"Staying Current\", \"line\": 516, \"offset\": 34498, \"end\": 34523}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Specifications and governance", "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution", "K:Prompt injection and malicious tools", "K:Vulnerability and secure code" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter09_patterns.tex" } }, { "id": "B094", "label": "Scheduling and Overnight Execution", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:System knowledge & memory", "T:Engineering structure", "T:Verification & review" ], "actual": [], "mapped": [ "C:C03", "C:C09", "C:C11" ], "printed": false, "summary": "L’esecuzione notturna richiede una sequenza controllata di job, non soltanto un timer.", "limitation": "L’ordine temporale deve essere verificato come dipendenza e recupero dagli errori; l’avvio schedulato non autorizza merge o deployment.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Scheduling and Overnight Execution", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter09b_scheduling.tex", "source_sha256": "ea1858d0de0f786aa947e1c33e48c1aaa43998e755ec2e84526a795cc92e9f76", "line_count": 933, "word_count_approx": 3890, "card_type": "chapter", "card_path": "cards/V04/B094.md", "in_short": "L’esecuzione notturna richiede una sequenza controllata di job, non soltanto un timer.", "summary": "Il file distingue revisione cumulativa e lavoro autonomo, descrive schedulatori per sistemi operativi diversi, esecuzione cloud e notifiche.", "limitation": "L’ordine temporale deve essere verificato come dipendenza e recupero dagli errori; l’avvio schedulato non autorizza merge o deployment.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:System knowledge & memory\", \"T:Engineering structure\", \"T:Verification & review\"]", "topics_json": "[\"K:Long-running agent execution\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C09\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Two-Job Sequence\", \"line\": 24, \"offset\": 1985, \"end\": 2015}, {\"level\": \"section\", \"title\": \"Scheduling on Linux (systemd/cron)\", \"line\": 381, \"offset\": 15906, \"end\": 15950}, {\"level\": \"section\", \"title\": \"Scheduling on Windows\", \"line\": 501, \"offset\": 19422, \"end\": 19453}, {\"level\": \"section\", \"title\": \"Monitoring and Notification\", \"line\": 733, \"offset\": 28592, \"end\": 28629}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Long-running agent execution", "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Specifications and governance", "K:Ownership and knowledge transfer" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter09b_scheduling.tex" } }, { "id": "B095", "label": "Multi-Agent Orchestration", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Engineering structure", "T:Verification & review", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C03", "C:C08", "C:C11" ], "printed": false, "summary": "Il multi-agente diventa utile quando isolamento e coordinamento risolvono un limite osservato del singolo loop.", "limitation": "Più agenti non implicano più indipendenza né throughput lineare; integrazione, merge e capacità di revisione restano vincoli.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Multi-Agent Orchestration", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter10_orchestration.tex", "source_sha256": "8f3c5c3e2f05c774aba371e5d85f132b2483dfde5357552d72c4ee3f7d95fe1c", "line_count": 744, "word_count_approx": 4996, "card_type": "chapter", "card_path": "cards/V04/B095.md", "in_short": "Il multi-agente diventa utile quando isolamento e coordinamento risolvono un limite osservato del singolo loop.", "summary": "Hat-based orchestration, planner-worker, passaggi umani e Gas Town sono confrontati come strutture di coordinamento con costi diversi.", "limitation": "Più agenti non implicano più indipendenza né throughput lineare; integrazione, merge e capacità di revisione restano vincoli.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Engineering structure\", \"T:Verification & review\", \"T:Code quality & security\"]", "topics_json": "[\"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\", \"K:AI as reviewer\", \"K:Automation bias and trust\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C08\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"When to Scale Beyond Single Loops\", \"line\": 20, \"offset\": 1643, \"end\": 1686}, {\"level\": \"section\", \"title\": \"Human-in-the-Loop Patterns\", \"line\": 277, \"offset\": 16282, \"end\": 16318}, {\"level\": \"section\", \"title\": \"Understanding Gas Town (Conceptually)\", \"line\": 500, \"offset\": 27527, \"end\": 27574}, {\"level\": \"section\", \"title\": \"The Frontier: Evolutionary Software\", \"line\": 624, \"offset\": 36444, \"end\": 36489}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution", "K:AI as reviewer", "K:Automation bias and trust", "K:Specifications and governance", "K:Ownership and knowledge transfer" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter10_orchestration.tex" } }, { "id": "B096", "label": "Safety, Security, and Risk Management", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Code quality & security", "T:What agents can do", "T:Engineering structure", "T:Economics", "T:Productivity", "T:Verification & review" ], "actual": [], "mapped": [ "C:C06", "C:C08", "C:C12" ], "printed": false, "summary": "La sicurezza dell’autonomia combina contenimento, limiti economici e una frontiera di revisione umana.", "limitation": "Una worktree isola modifiche Git, non processi, rete o credenziali. Le prescrizioni del volume non sono una certificazione della configurazione effettiva.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Safety, Security, and Risk Management", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter11_safety.tex", "source_sha256": "ba3475482cee7bf1589b6b5bf5cdef2e6b0b5b87a3d264dffc5066762579d362", "line_count": 744, "word_count_approx": 4753, "card_type": "chapter", "card_path": "cards/V04/B096.md", "in_short": "La sicurezza dell’autonomia combina contenimento, limiti economici e una frontiera di revisione umana.", "summary": "Sandboxing, prevenzione di azioni distruttive, overbaking e gestione dei costi convergono in una checklist operativa.", "limitation": "Una worktree isola modifiche Git, non processi, rete o credenziali. Le prescrizioni del volume non sono una certificazione della configurazione effettiva.", "primary_theme": "T:Code quality & security", "streams_json": "[\"T:Code quality & security\", \"T:What agents can do\", \"T:Engineering structure\", \"T:Economics\", \"T:Productivity\", \"T:Verification & review\"]", "topics_json": "[\"K:Prompt injection and malicious tools\", \"K:Vulnerability and secure code\", \"K:Economic costs and return\", \"K:Output versus shipped value\", \"K:Review capacity and maintainers\", \"K:AI as reviewer\", \"K:Automation bias and trust\"]", "editorial_chapters_json": "[\"C:C06\", \"C:C08\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Security Reality\", \"line\": 20, \"offset\": 1561, \"end\": 1591}, {\"level\": \"section\", \"title\": \"Preventing Destructive Actions\", \"line\": 295, \"offset\": 18722, \"end\": 18762}, {\"level\": \"section\", \"title\": \"Cost Management\", \"line\": 475, \"offset\": 28123, \"end\": 28148}, {\"level\": \"section\", \"title\": \"Putting It All Together: A Safety Checklist\", \"line\": 667, \"offset\": 38325, \"end\": 38378}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Prompt injection and malicious tools", "K:Vulnerability and secure code", "K:Economic costs and return", "K:Output versus shipped value", "K:Review capacity and maintainers", "K:AI as reviewer", "K:Automation bias and trust" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/chapters/chapter11_safety.tex" } }, { "id": "B097", "label": "Complete Implementation Guide", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Engineering structure", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C03", "C:C06", "C:C07" ], "printed": false, "summary": "La guida d’implementazione rende il pattern eseguibile mediante script, piani, prompt e checklist.", "limitation": "L’etichetta production-ready del sorgente non è stata convalidata da esecuzione o audit di sicurezza in questo lavoro; il codice rimane materiale del volume.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Complete Implementation Guide", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/appendices/appendix_a_ralph.tex", "source_sha256": "399ba75653a266acf8f26b77478d5f1c1f6e545cc0acd1d2851d7c142eab9b2d", "line_count": 1492, "word_count_approx": 4764, "card_type": "appendix", "card_path": "cards/V04/B097.md", "in_short": "La guida d’implementazione rende il pattern eseguibile mediante script, piani, prompt e checklist.", "summary": "Il file offre verifica dell’ambiente, script Ralph, template di memoria e pianificazione, istruzioni di avvio e prerequisiti per l’uso operativo.", "limitation": "L’etichetta production-ready del sorgente non è stata convalidata da esecuzione o audit di sicurezza in questo lavoro; il codice rimane materiale del volume.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Engineering structure\", \"T:Code quality & security\"]", "topics_json": "[\"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\", \"K:Prompt injection and malicious tools\", \"K:Vulnerability and secure code\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C06\", \"C:C07\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Environment Verification\", \"line\": 14, \"offset\": 771, \"end\": 805}, {\"level\": \"section\", \"title\": \"Plan File Templates\", \"line\": 1153, \"offset\": 28975, \"end\": 29004}, {\"level\": \"section\", \"title\": \"Planning Mode Prompt\", \"line\": 1332, \"offset\": 32432, \"end\": 32462}, {\"level\": \"section\", \"title\": \"Checklist for Production Use\", \"line\": 1474, \"offset\": 35514, \"end\": 35552}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles", "K:Prompt injection and malicious tools", "K:Vulnerability and secure code", "K:Evaluation methodology and baselines" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/appendices/appendix_a_ralph.tex" } }, { "id": "B098", "label": "Tool Reference", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Productivity", "T:Skills & learning", "T:What agents can do", "T:Verification & review" ], "actual": [], "mapped": [ "C:C03", "C:C11" ], "printed": false, "summary": "La reference confronta implementazioni Ralph, configurazioni e percorsi di migrazione.", "limitation": "Repository, versioni e funzionalità sono uno snapshot del libro, non una verifica dello stato attuale degli strumenti.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Tool Reference", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/appendices/appendix_b_tools.tex", "source_sha256": "f948d7baa29121873299c3d9302770c76dac6ef9593799a342a5d9000b869efe", "line_count": 1103, "word_count_approx": 3578, "card_type": "appendix", "card_path": "cards/V04/B098.md", "in_short": "La reference confronta implementazioni Ralph, configurazioni e percorsi di migrazione.", "summary": "Le varianti snarktank, plugin Anthropic, frankbria e orchestrator vengono organizzate per selezione, compatibilità e diagnosi specifica.", "limitation": "Repository, versioni e funzionalità sono uno snapshot del libro, non una verifica dello stato attuale degli strumenti.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Productivity\", \"T:Skills & learning\", \"T:What agents can do\", \"T:Verification & review\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\", \"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Implementation Comparison\", \"line\": 14, \"offset\": 807, \"end\": 842}, {\"level\": \"section\", \"title\": \"frankbria/ralph-claude-code\", \"line\": 266, \"offset\": 10415, \"end\": 10452}, {\"level\": \"section\", \"title\": \"Migration Guide\", \"line\": 734, \"offset\": 24795, \"end\": 24820}, {\"level\": \"section\", \"title\": \"Version Compatibility\", \"line\": 1075, \"offset\": 34010, \"end\": 34041}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Evaluation methodology and baselines", "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/appendices/appendix_b_tools.tex" } }, { "id": "B099", "label": "Troubleshooting and Debugging", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Engineering structure", "T:Economics", "T:Productivity" ], "actual": [], "mapped": [ "C:C03", "C:C07", "C:C12" ], "printed": false, "summary": "La diagnosi parte dai sintomi del loop e ricostruisce cause, stato e costi.", "limitation": "Le ricette diagnostiche sono proposte operative; l’efficacia e la sicurezza dei comandi dipendono dall’ambiente e non sono state provate qui.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Troubleshooting and Debugging", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/appendices/appendix_c_troubleshooting.tex", "source_sha256": "5d46441c9f5b502515a9b7baa5fc59e303cdde7b809eeaee04e627c6d6000ca3", "line_count": 1454, "word_count_approx": 5267, "card_type": "appendix", "card_path": "cards/V04/B099.md", "in_short": "La diagnosi parte dai sintomi del loop e ricostruisce cause, stato e costi.", "summary": "Uscita precoce, mancato arresto, gate falliti, duplicazioni e consumi fuori controllo vengono collegati a log, Git, recupero e procedure d’emergenza.", "limitation": "Le ricette diagnostiche sono proposte operative; l’efficacia e la sicurezza dei comandi dipendono dall’ambiente e non sono state provate qui.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Engineering structure\", \"T:Economics\", \"T:Productivity\"]", "topics_json": "[\"K:Agent failure and escalation\", \"K:Evaluation methodology and baselines\", \"K:Long-running agent execution\", \"K:Independent verification and oracles\", \"K:Economic costs and return\", \"K:Output versus shipped value\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C07\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Loop Exits Early\", \"line\": 16, \"offset\": 911, \"end\": 937}, {\"level\": \"section\", \"title\": \"Agent Gets Stuck on Same Issue\", \"line\": 447, \"offset\": 14722, \"end\": 14762}, {\"level\": \"section\", \"title\": \"Performance Debugging\", \"line\": 1009, \"offset\": 31903, \"end\": 31934}, {\"level\": \"section\", \"title\": \"Appendix Summary\", \"line\": 1438, \"offset\": 44444, \"end\": 44471}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Agent failure and escalation", "K:Evaluation methodology and baselines", "K:Long-running agent execution", "K:Independent verification and oracles", "K:Economic costs and return", "K:Output versus shipped value" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/appendices/appendix_c_troubleshooting.tex" } }, { "id": "B100", "label": "Ralph Loops with VS Code and GitHub Copilot", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Engineering structure", "T:Code quality & security", "T:Productivity", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C03", "C:C06" ], "printed": false, "summary": "Ralph viene adattato a Copilot tramite estensioni, CLI, subagenti o iterazione manuale.", "limitation": "Supporto dei modelli, estensioni e opzioni CLI possono essere cambiati; i profili di autorizzazione vanno letti come descrizione storica del volume.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Ralph Loops with VS Code and GitHub Copilot", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/appendices/appendix_d_copilot.tex", "source_sha256": "7e0010a74aa385fc08a8e847aac932a2d656978a938a3cfc4a013b55afb63728", "line_count": 778, "word_count_approx": 4256, "card_type": "appendix", "card_path": "cards/V04/B100.md", "in_short": "Ralph viene adattato a Copilot tramite estensioni, CLI, subagenti o iterazione manuale.", "summary": "Il capitolo confronta quattro approcci con permessi, modelli, formato del piano, contesto e integrazione con lo scheduling.", "limitation": "Supporto dei modelli, estensioni e opzioni CLI possono essere cambiati; i profili di autorizzazione vanno letti come descrizione storica del volume.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Engineering structure\", \"T:Code quality & security\", \"T:Productivity\", \"T:Skills & learning\"]", "topics_json": "[\"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\", \"K:Prompt injection and malicious tools\", \"K:Vulnerability and secure code\", \"K:Developer workflow and adoption\", \"K:Evaluation methodology and baselines\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C06\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Understanding the Differences\", \"line\": 14, \"offset\": 1007, \"end\": 1046}, {\"level\": \"section\", \"title\": \"Approach 2: GitHub Copilot CLI (Recommended)\", \"line\": 257, \"offset\": 11855, \"end\": 11909}, {\"level\": \"section\", \"title\": \"Approach 3: Subagent Orchestration in VS Code Chat\", \"line\": 573, \"offset\": 24987, \"end\": 25047}, {\"level\": \"section\", \"title\": \"Choosing Your Approach\", \"line\": 693, \"offset\": 31279, \"end\": 31311}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles", "K:Prompt injection and malicious tools", "K:Vulnerability and secure code", "K:Developer workflow and adoption", "K:Evaluation methodology and baselines" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/appendices/appendix_d_copilot.tex" } }, { "id": "B101", "label": "References and Sources", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do", "T:Verification & review" ], "actual": [], "mapped": [ "C:C01", "C:C03" ], "printed": false, "summary": "L’apparato di fonti rende esplicita la genealogia del metodo autonomo.", "limitation": "Le note «Referenced in» sono attribuzioni dell’apparato, non un riscontro automatico nel testo di ogni capitolo; gli URL dichiarati verificati nel febbraio 2026 non sono stati ricontrollati qui.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "References and Sources", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V04", "volume_title": "Non-Deterministic Autonomous Coding Agents", "source_path": "ai-auto_eng/AI_AUTO_ENG/bibliography.tex", "source_sha256": "a86fb72afc5f33dca04aec60b1a12a9d687dda31e73fd6cdf499a414e093aed1", "line_count": 436, "word_count_approx": 1405, "card_type": "bibliographic_apparatus", "card_path": "cards/V04/B101.md", "in_short": "L’apparato di fonti rende esplicita la genealogia del metodo autonomo.", "summary": "Libri, persone, articoli, repository, video, casi e concetti attribuiti sono raccolti con indicazioni dei capitoli interessati.", "limitation": "Le note «Referenced in» sono attribuzioni dell’apparato, non un riscontro automatico nel testo di ogni capitolo; gli URL dichiarati verificati nel febbraio 2026 non sono stati ricontrollati qui.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\", \"T:Verification & review\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Human-agent collaboration\", \"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C03\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Books\", \"line\": 17, \"offset\": 343, \"end\": 359}, {\"level\": \"section\", \"title\": \"GitHub Repositories --- Ralph Implementations\", \"line\": 158, \"offset\": 6204, \"end\": 6260}, {\"level\": \"section\", \"title\": \"Video Series and Talks\", \"line\": 323, \"offset\": 12221, \"end\": 12254}, {\"level\": \"section\", \"title\": \"Concepts and Frameworks Attributed\", \"line\": 401, \"offset\": 15185, \"end\": 15230}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Human-agent collaboration", "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles" ], "volumes": [ "V04" ], "book": { "volume": "V04", "title": "Non-Deterministic Autonomous Coding Agents", "subtitle": "Building Self-Improving Systems That Ship While You Sleep", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "ai-auto_eng/AI_AUTO_ENG/bibliography.tex" } }, { "id": "B102", "label": "Avvertenze editoriali e limiti della guida", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:What agents can do" ], "actual": [], "mapped": [ "C:C01" ], "printed": false, "summary": "Le pagine iniziali qualificano finalità, responsabilità e natura della sintesi sui loop.", "limitation": "Il frontmatter è contesto di lettura e non fornisce evidenza empirica aggiuntiva alle fonti o ai capitoli.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Avvertenze editoriali e limiti della guida", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V05", "volume_title": "Non-Deterministic Loop Engineering", "source_path": "NDLE/frontmatter.tex", "source_sha256": "a92fea666811437893828facf85543449e9c4e434d2e078ef86531a0c842dca7", "line_count": 128, "word_count_approx": 456, "card_type": "reading_guide", "card_path": "cards/V05/B102.md", "in_short": "Le pagine iniziali qualificano finalità, responsabilità e natura della sintesi sui loop.", "summary": "Le avvertenze editoriali aiutano a distinguere il libro come guida ingegneristica da una politica aziendale o da una garanzia di risultato.", "limitation": "Il frontmatter è contesto di lettura e non fornisce evidenza empirica aggiuntiva alle fonti o ai capitoli.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:What agents can do\"]", "topics_json": "[\"K:Developer workflow and adoption\", \"K:Human-agent collaboration\"]", "editorial_chapters_json": "[\"C:C01\"]", "evidence_anchors_json": "[{\"title\": \"Avvertenze editoriali e limiti della guida\", \"line\": 1}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Developer workflow and adoption", "K:Human-agent collaboration" ], "volumes": [ "V05" ], "book": { "volume": "V05", "title": "Non-Deterministic Loop Engineering", "subtitle": "Design Reliable Agentic AI Systems Beyond the Prompt", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "NDLE/frontmatter.tex" } }, { "id": "B103", "label": "Summary", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Engineering structure", "T:Economics", "T:Productivity" ], "actual": [], "mapped": [ "C:C01", "C:C03", "C:C11", "C:C12" ], "printed": false, "summary": "Il sommario introduce il loop come sistema governato, dal modello concettuale all’esecuzione locale.", "limitation": "La sintesi organizza il volume; non va conteggiata come conferma indipendente degli stessi concetti nei capitoli estesi.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Summary", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V05", "volume_title": "Non-Deterministic Loop Engineering", "source_path": "NDLE/summary_v2.tex", "source_sha256": "a2c61f272e6afb5d3f3efedf51856604bedc834f0de2235d548ae7ee89749539", "line_count": 38, "word_count_approx": 348, "card_type": "synthesis", "card_path": "cards/V05/B103.md", "in_short": "Il sommario introduce il loop come sistema governato, dal modello concettuale all’esecuzione locale.", "summary": "La panoramica collega critica, organizzazione, economia, Ralph, runtime locale e scelta del meccanismo minimo sufficiente.", "limitation": "La sintesi organizza il volume; non va conteggiata come conferma indipendente degli stessi concetti nei capitoli estesi.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Engineering structure\", \"T:Economics\", \"T:Productivity\"]", "topics_json": "[\"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\", \"K:Economic costs and return\", \"K:Output versus shipped value\"]", "editorial_chapters_json": "[\"C:C01\", \"C:C03\", \"C:C11\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"How This Book Is Organized\", \"line\": 26, \"offset\": 1515, \"end\": 1552}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles", "K:Specifications and governance", "K:Ownership and knowledge transfer", "K:Economic costs and return", "K:Output versus shipped value" ], "volumes": [ "V05" ], "book": { "volume": "V05", "title": "Non-Deterministic Loop Engineering", "subtitle": "Design Reliable Agentic AI Systems Beyond the Prompt", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "NDLE/summary_v2.tex" } }, { "id": "B104", "label": "Non-Deterministic Loop Engineering: A Critical Assessment", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Engineering structure", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C03", "C:C07", "C:C09", "C:C11" ], "printed": false, "summary": "Il loop affidabile emerge dall’interazione fra contesto, harness, evidenza e autorità.", "limitation": "I resoconti practitioner e le tassonomie di autonomia non stabiliscono frequenze di fallimento; separare ciò che è riportato dalle inferenze e dalle proposte dell’autore.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Non-Deterministic Loop Engineering: A Critical Assessment", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V05", "volume_title": "Non-Deterministic Loop Engineering", "source_path": "NDLE/new_chapters/chapter01_foundations.tex", "source_sha256": "bc900422e5a5ba75bc505fe8bc5cb5062604be18ffbad112f277c8367bd2f7d8", "line_count": 1549, "word_count_approx": 8059, "card_type": "chapter", "card_path": "cards/V05/B104.md", "in_short": "Il loop affidabile emerge dall’interazione fra contesto, harness, evidenza e autorità.", "summary": "Il capitolo ricostruisce la genealogia prompt–loop, distingue worker ed evaluator, presenta livelli di delega e critica rischi sistemici e non-determinismo.", "limitation": "I resoconti practitioner e le tassonomie di autonomia non stabiliscono frequenze di fallimento; separare ciò che è riportato dalle inferenze e dalle proposte dell’autore.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Engineering structure\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\", \"K:AI as reviewer\", \"K:Automation bias and trust\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Dependencies and agent supply chain\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C07\", \"C:C09\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Introduction: The Epistemological Transition in Software Development\", \"line\": 12, \"offset\": 795, \"end\": 873}, {\"level\": \"section\", \"title\": \"From /loop to /goal : Native Loop Primitives\", \"line\": 566, \"offset\": 24529, \"end\": 24600}, {\"level\": \"section\", \"title\": \"Empirical Landscape: What Practitioners Actually Report\", \"line\": 1071, \"offset\": 40241, \"end\": 40306}, {\"level\": \"section\", \"title\": \"The Joint System: Context, Harness, Evidence, and Policy\", \"line\": 1483, \"offset\": 59593, \"end\": 59659}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles", "K:AI as reviewer", "K:Automation bias and trust", "K:Specifications and governance", "K:Ownership and knowledge transfer", "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Dependencies and agent supply chain" ], "volumes": [ "V05" ], "book": { "volume": "V05", "title": "Non-Deterministic Loop Engineering", "subtitle": "Design Reliable Agentic AI Systems Beyond the Prompt", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "NDLE/new_chapters/chapter01_foundations.tex" } }, { "id": "B105", "label": "The Organization Becomes a Loop", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Engineering structure", "T:Verification & review", "T:Jobs & roles", "T:Skills & learning", "T:System knowledge & memory", "T:What agents can do", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C08", "C:C09", "C:C10", "C:C11" ], "printed": false, "summary": "L’organizzazione deve assegnare responsabilità, capacità di verifica e infrastruttura di conoscenza al lavoro dei loop.", "limitation": "Le «cinque leggi» sono principi proposti, non leggi empiriche dimostrate; misurare handoff, capacità di revisione, mantenimento delle competenze e risultati di un pilot.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Organization Becomes a Loop", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V05", "volume_title": "Non-Deterministic Loop Engineering", "source_path": "NDLE/new_chapters/chapter02_organization.tex", "source_sha256": "b8127eb334612ffa241bb6590f32fb0ac494bb4270e54bd9bd4f9ab6391f1f12", "line_count": 818, "word_count_approx": 6092, "card_type": "chapter", "card_path": "cards/V05/B105.md", "in_short": "L’organizzazione deve assegnare responsabilità, capacità di verifica e infrastruttura di conoscenza al lavoro dei loop.", "summary": "Ruoli, modelli operativi, metriche, governance nel runtime e roadmap di transizione descrivono come ridisegnare team e management.", "limitation": "Le «cinque leggi» sono principi proposti, non leggi empiriche dimostrate; misurare handoff, capacità di revisione, mantenimento delle competenze e risultati di un pilot.", "primary_theme": "T:Engineering structure", "streams_json": "[\"T:Engineering structure\", \"T:Verification & review\", \"T:Jobs & roles\", \"T:Skills & learning\", \"T:System knowledge & memory\", \"T:What agents can do\", \"T:Code quality & security\"]", "topics_json": "[\"K:Specifications and governance\", \"K:Ownership and knowledge transfer\", \"K:Employment and early-career roles\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Review capacity and maintainers\", \"K:AI as reviewer\", \"K:Automation bias and trust\"]", "editorial_chapters_json": "[\"C:C08\", \"C:C09\", \"C:C10\", \"C:C11\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"From a New Tool to a New Institution\", \"line\": 26, \"offset\": 3160, \"end\": 3206}, {\"level\": \"section\", \"title\": \"Three Enterprise Operating Models\", \"line\": 339, \"offset\": 20886, \"end\": 20929}, {\"level\": \"section\", \"title\": \"The Workforce Impact: Rebundling, Not Simple Replacement\", \"line\": 588, \"offset\": 32838, \"end\": 32904}, {\"level\": \"section\", \"title\": \"The Leadership Paradox\", \"line\": 759, \"offset\": 45921, \"end\": 45953}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Specifications and governance", "K:Ownership and knowledge transfer", "K:Employment and early-career roles", "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Review capacity and maintainers", "K:AI as reviewer", "K:Automation bias and trust" ], "volumes": [ "V05" ], "book": { "volume": "V05", "title": "Non-Deterministic Loop Engineering", "subtitle": "Design Reliable Agentic AI Systems Beyond the Prompt", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "NDLE/new_chapters/chapter02_organization.tex" } }, { "id": "B106", "label": "The Economics of Loop Engineering", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Economics", "T:Productivity", "T:Verification & review", "T:Jobs & roles", "T:Skills & learning" ], "actual": [], "mapped": [ "C:C02", "C:C11", "C:C12" ], "printed": false, "summary": "L’unità economica diventa l’esito verificato, non il token economico o la velocità del singolo contributore.", "limitation": "Scenari ed esempi analitici non sono contabilità produttiva misurata. Parametri di accettazione, tempi umani e perdite attese richiedono calibrazione indipendente.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Economics of Loop Engineering", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V05", "volume_title": "Non-Deterministic Loop Engineering", "source_path": "NDLE/new_chapters/chapter03_economics.tex", "source_sha256": "056669d326018cda56fff9a98d80302704953af32d4cbacf92673f89ad13d180", "line_count": 1149, "word_count_approx": 6895, "card_type": "chapter", "card_path": "cards/V05/B106.md", "in_short": "L’unità economica diventa l’esito verificato, non il token economico o la velocità del singolo contributore.", "summary": "Costi di esecuzione, verifica, rilavorazione e coordinamento sostengono scenari organizzativi e un caso di manutenzione in ambiente regolamentato.", "limitation": "Scenari ed esempi analitici non sono contabilità produttiva misurata. Parametri di accettazione, tempi umani e perdite attese richiedono calibrazione indipendente.", "primary_theme": "T:Economics", "streams_json": "[\"T:Economics\", \"T:Productivity\", \"T:Verification & review\", \"T:Jobs & roles\", \"T:Skills & learning\"]", "topics_json": "[\"K:Economic costs and return\", \"K:Output versus shipped value\", \"K:Evaluation methodology and baselines\", \"K:Controlled productivity evidence\", \"K:Employment and early-career roles\", \"K:Ownership and knowledge transfer\"]", "editorial_chapters_json": "[\"C:C02\", \"C:C11\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Introduction: From Cheap Assistance to Variable-Cost Production\", \"line\": 12, \"offset\": 699, \"end\": 772}, {\"level\": \"section\", \"title\": \"The Organization: Local Speed Is Not System Throughput\", \"line\": 455, \"offset\": 22641, \"end\": 22705}, {\"level\": \"section\", \"title\": \"Four Economic Failure Modes\", \"line\": 836, \"offset\": 38279, \"end\": 38316}, {\"level\": \"section\", \"title\": \"End-to-End Case Study: A Regulated Maintenance Loop\", \"line\": 925, \"offset\": 45884, \"end\": 45945}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Economic costs and return", "K:Output versus shipped value", "K:Evaluation methodology and baselines", "K:Controlled productivity evidence", "K:Employment and early-career roles", "K:Ownership and knowledge transfer" ], "volumes": [ "V05" ], "book": { "volume": "V05", "title": "Non-Deterministic Loop Engineering", "subtitle": "Design Reliable Agentic AI Systems Beyond the Prompt", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "NDLE/new_chapters/chapter03_economics.tex" } }, { "id": "B107", "label": "The Ralph Loop: Persistent Progress Through Fresh Context", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Engineering structure", "T:System knowledge & memory", "T:Code quality & security" ], "actual": [], "mapped": [ "C:C03", "C:C05", "C:C06", "C:C07", "C:C09" ], "printed": false, "summary": "Ralph conserva il progresso fuori dalla sessione e tratta la ripetizione come una politica limitata e verificabile.", "limitation": "La persistenza non assicura convergenza; prova di completamento, autorizzazione e rollback devono risiedere fuori dalle dichiarazioni del worker.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "The Ralph Loop: Persistent Progress Through Fresh Context", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V05", "volume_title": "Non-Deterministic Loop Engineering", "source_path": "NDLE/new_chapters/chapter04_ralph_loop.tex", "source_sha256": "aa58f0631ba2700b7664babd919b1e842cf17557b9d86fd374bab6fac5fc0636", "line_count": 1324, "word_count_approx": 6157, "card_type": "chapter", "card_path": "cards/V05/B107.md", "in_short": "Ralph conserva il progresso fuori dalla sessione e tratta la ripetizione come una politica limitata e verificabile.", "summary": "Il capitolo confronta loop Bash, stop hook e modalità manuale; aggiunge unità di lavoro, budget, escalation, manutenibilità e criteri di scelta.", "limitation": "La persistenza non assicura convergenza; prova di completamento, autorizzazione e rollback devono risiedere fuori dalle dichiarazioni del worker.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Engineering structure\", \"T:System knowledge & memory\", \"T:Code quality & security\"]", "topics_json": "[\"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Prompt injection and malicious tools\", \"K:Vulnerability and secure code\", \"K:Architecture and modularity\", \"K:Technical debt and complexity\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C05\", \"C:C06\", \"C:C07\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"Why the Ralph Loop Matters\", \"line\": 36, \"offset\": 2668, \"end\": 2704}, {\"level\": \"section\", \"title\": \"Implementation Pattern II: The Stop-Hook Loop\", \"line\": 545, \"offset\": 21272, \"end\": 21327}, {\"level\": \"section\", \"title\": \"What Ralph Does Not Solve\", \"line\": 1117, \"offset\": 38204, \"end\": 38239}, {\"level\": \"section\", \"title\": \"Where the Minimal Loop Begins to Strain\", \"line\": 1285, \"offset\": 46632, \"end\": 46681}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles", "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Prompt injection and malicious tools", "K:Vulnerability and secure code", "K:Architecture and modularity", "K:Technical debt and complexity" ], "volumes": [ "V05" ], "book": { "volume": "V05", "title": "Non-Deterministic Loop Engineering", "subtitle": "Design Reliable Agentic AI Systems Beyond the Prompt", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "NDLE/new_chapters/chapter04_ralph_loop.tex" } }, { "id": "B108", "label": "Building a Full Local AI Loop", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Economics", "T:What agents can do", "T:Code quality & security", "T:Engineering structure", "T:Verification & review", "T:System knowledge & memory", "T:Productivity" ], "actual": [], "mapped": [ "C:C03", "C:C06", "C:C07", "C:C09", "C:C12" ], "printed": false, "summary": "Un loop locale completo include runtime controllato, intenti durevoli, verifica e telemetria economica.", "limitation": "La convenienza locale dipende da utilizzo, energia, ammortamento e tasso di accettazione. Installazioni e benchmark citati non sono stati rieseguiti; il titolo Start and Verify the Local Services ricorre due volte.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Building a Full Local AI Loop", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V05", "volume_title": "Non-Deterministic Loop Engineering", "source_path": "NDLE/new_chapters/chapter05_local_ai_loop.tex", "source_sha256": "702b6ef3084dd35d95c15d7a44021b61a3ef7c556325541e0e433c983a7b9cb2", "line_count": 1782, "word_count_approx": 10333, "card_type": "chapter", "card_path": "cards/V05/B108.md", "in_short": "Un loop locale completo include runtime controllato, intenti durevoli, verifica e telemetria economica.", "summary": "Il percorso attraversa dimensionamento, energia e costo totale, installazione, scelta dell’executor, retry e resume, policy, osservabilità e gestione della deriva.", "limitation": "La convenienza locale dipende da utilizzo, energia, ammortamento e tasso di accettazione. Installazioni e benchmark citati non sono stati rieseguiti; il titolo Start and Verify the Local Services ricorre due volte.", "primary_theme": "T:Economics", "streams_json": "[\"T:Economics\", \"T:What agents can do\", \"T:Code quality & security\", \"T:Engineering structure\", \"T:Verification & review\", \"T:System knowledge & memory\", \"T:Productivity\"]", "topics_json": "[\"K:Economic costs and return\", \"K:Dependencies and agent supply chain\", \"K:Long-running agent execution\", \"K:Prompt injection and malicious tools\", \"K:Vulnerability and secure code\", \"K:Independent verification and oracles\", \"K:Evaluation methodology and baselines\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\", \"K:Output versus shipped value\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C06\", \"C:C07\", \"C:C09\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"The Decisive Shift: From Local Inference to Local Engineering\", \"line\": 12, \"offset\": 650, \"end\": 721}, {\"level\": \"section\", \"title\": \"Configuration as a Control Plane\", \"line\": 816, \"offset\": 34881, \"end\": 34923}, {\"level\": \"section\", \"title\": \"Loop Modes and Human Authority\", \"line\": 1345, \"offset\": 59249, \"end\": 59289}, {\"level\": \"section\", \"title\": \"The Durable Engineering Principles\", \"line\": 1725, \"offset\": 81951, \"end\": 81995}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Economic costs and return", "K:Dependencies and agent supply chain", "K:Long-running agent execution", "K:Prompt injection and malicious tools", "K:Vulnerability and secure code", "K:Independent verification and oracles", "K:Evaluation methodology and baselines", "K:Agent context and persistent memory", "K:System understanding and rationale", "K:Output versus shipped value" ], "volumes": [ "V05" ], "book": { "volume": "V05", "title": "Non-Deterministic Loop Engineering", "subtitle": "Design Reliable Agentic AI Systems Beyond the Prompt", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "NDLE/new_chapters/chapter05_local_ai_loop.tex" } }, { "id": "B109", "label": "Choosing the Right Loop", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:Verification & review", "T:Code quality & security", "T:What agents can do", "T:Engineering structure", "T:Economics", "T:Productivity" ], "actual": [], "mapped": [ "C:C03", "C:C07", "C:C08", "C:C11", "C:C12" ], "printed": false, "summary": "Si sceglie il meccanismo minimo che affronta un fallimento osservato e dispone di un segnale esterno verificabile.", "limitation": "La trasferibilità fra codice e assistenza clienti è una proposta di progetto; successo operativo, indipendenza del controllo e costo per esito richiedono valutazione nel dominio d’arrivo.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Choosing the Right Loop", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V05", "volume_title": "Non-Deterministic Loop Engineering", "source_path": "NDLE/new_chapters/chapter06_choosing_the_right_loop.tex", "source_sha256": "a061daf1e97a16c322a46cf9724dd0d92531718b57a038decdf521c6103f68b9", "line_count": 1580, "word_count_approx": 10591, "card_type": "chapter", "card_path": "cards/V05/B109.md", "in_short": "Si sceglie il meccanismo minimo che affronta un fallimento osservato e dispone di un segnale esterno verificabile.", "summary": "Famiglie di tecniche, scala dell’evidenza, contratto del loop e decision framework culminano in un caso non-code di supporto clienti.", "limitation": "La trasferibilità fra codice e assistenza clienti è una proposta di progetto; successo operativo, indipendenza del controllo e costo per esito richiedono valutazione nel dominio d’arrivo.", "primary_theme": "T:Verification & review", "streams_json": "[\"T:Verification & review\", \"T:Code quality & security\", \"T:What agents can do\", \"T:Engineering structure\", \"T:Economics\", \"T:Productivity\"]", "topics_json": "[\"K:Independent verification and oracles\", \"K:Evaluation methodology and baselines\", \"K:Agent failure and escalation\", \"K:Human-agent collaboration\", \"K:Review capacity and maintainers\", \"K:Long-running agent execution\", \"K:Economic costs and return\", \"K:Output versus shipped value\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C07\", \"C:C08\", \"C:C11\", \"C:C12\"]", "evidence_anchors_json": "[{\"level\": \"section\", \"title\": \"From a Complete Architecture to a Selective Operating Model\", \"line\": 12, \"offset\": 704, \"end\": 773}, {\"level\": \"section\", \"title\": \"Technique Family IV: Budget, Observability, and Guardrails\", \"line\": 628, \"offset\": 32214, \"end\": 32282}, {\"level\": \"section\", \"title\": \"A Build Sequence That Minimizes Accidental Complexity\", \"line\": 1163, \"offset\": 53964, \"end\": 54027}, {\"level\": \"section\", \"title\": \"Conclusion: Engineering Accountable Autonomy\", \"line\": 1565, \"offset\": 87266, \"end\": 87320}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Independent verification and oracles", "K:Evaluation methodology and baselines", "K:Agent failure and escalation", "K:Human-agent collaboration", "K:Review capacity and maintainers", "K:Long-running agent execution", "K:Economic costs and return", "K:Output versus shipped value" ], "volumes": [ "V05" ], "book": { "volume": "V05", "title": "Non-Deterministic Loop Engineering", "subtitle": "Design Reliable Agentic AI Systems Beyond the Prompt", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "NDLE/new_chapters/chapter06_choosing_the_right_loop.tex" } }, { "id": "B110", "label": "References — 85 voci bibliografiche", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Engineering structure", "T:Economics", "T:Productivity" ], "actual": [], "mapped": [ "C:C03", "C:C11", "C:C12" ], "printed": false, "summary": "Le 85 voci bibliografiche raccolgono fonti practitioner, ricerca, strumenti e casi richiamati nel volume.", "limitation": "Identità bibliografica non significa supporto di ogni claim: fonti ripubblicate, articoli derivati e casi vendor non sono repliche indipendenti.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "References — 85 voci bibliografiche", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V05", "volume_title": "Non-Deterministic Loop Engineering", "source_path": "NDLE/references.tex", "source_sha256": "0f40753fccc8f143d768071a179a417deeeaa4afeb0698917b4ecd3e08987692", "line_count": 496, "word_count_approx": 1777, "card_type": "bibliographic_apparatus", "card_path": "cards/V05/B110.md", "in_short": "Le 85 voci bibliografiche raccolgono fonti practitioner, ricerca, strumenti e casi richiamati nel volume.", "summary": "Le chiavi bibitem consentono di risolvere citazioni esplicite dei capitoli e separare la presenza in bibliografia dall’uso effettivo di una fonte.", "limitation": "Identità bibliografica non significa supporto di ogni claim: fonti ripubblicate, articoli derivati e casi vendor non sono repliche indipendenti.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Engineering structure\", \"T:Economics\", \"T:Productivity\"]", "topics_json": "[\"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\", \"K:Specifications and governance\", \"K:Ownership and knowledge transfer\", \"K:Economic costs and return\", \"K:Output versus shipped value\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C11\", \"C:C12\"]", "evidence_anchors_json": "[{\"title\": \"References — 85 voci bibliografiche\", \"line\": 1}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles", "K:Specifications and governance", "K:Ownership and knowledge transfer", "K:Economic costs and return", "K:Output versus shipped value" ], "volumes": [ "V05" ], "book": { "volume": "V05", "title": "Non-Deterministic Loop Engineering", "subtitle": "Design Reliable Agentic AI Systems Beyond the Prompt", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "NDLE/references.tex" } }, { "id": "B111", "label": "Glossary", "kind": "author_tex_section", "collection": "B", "category": null, "streams": [ "T:What agents can do", "T:Verification & review", "T:Engineering structure", "T:Code quality & security", "T:System knowledge & memory" ], "actual": [], "mapped": [ "C:C03", "C:C07", "C:C09" ], "printed": false, "summary": "Il glossario stabilizza il significato di loop limitato, stato durevole, backpressure ed evidenza.", "limitation": "Una definizione è un vincolo terminologico interno, non una misura validata né un consenso universale della disciplina.", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "authors": "Enrico Papalini", "year": "", "attributes": { "label": "Glossary", "kind": "author_tex_section", "node_class": "ND_INFORMATIONAL_TEX", "collection": "B", "volume_id": "V05", "volume_title": "Non-Deterministic Loop Engineering", "source_path": "NDLE/glossary.tex", "source_sha256": "39dbffdfcbff6363995d8c818eeaefeae1014ac01c26f98c3b5ab99f05ba86ef", "line_count": 110, "word_count_approx": 1148, "card_type": "glossary", "card_path": "cards/V05/B111.md", "in_short": "Il glossario stabilizza il significato di loop limitato, stato durevole, backpressure ed evidenza.", "summary": "Le definizioni separano segnale di completamento e prova, accettazione e stato provvisorio, capacità del modello e controlli del sistema.", "limitation": "Una definizione è un vincolo terminologico interno, non una misura validata né un consenso universale della disciplina.", "primary_theme": "T:What agents can do", "streams_json": "[\"T:What agents can do\", \"T:Verification & review\", \"T:Engineering structure\", \"T:Code quality & security\", \"T:System knowledge & memory\"]", "topics_json": "[\"K:Long-running agent execution\", \"K:Agent failure and escalation\", \"K:Independent verification and oracles\", \"K:Evaluation methodology and baselines\", \"K:Agent context and persistent memory\", \"K:System understanding and rationale\"]", "editorial_chapters_json": "[\"C:C03\", \"C:C07\", \"C:C09\"]", "evidence_anchors_json": "[{\"level\": \"chapter\", \"title\": \"Glossary\", \"line\": 1, \"offset\": 0, \"end\": 19}]", "author": "Enrico Papalini", "evidence_class": "Author-authored engineering synthesis, guidance or reference apparatus", "review_scope": "File-level content mapping and citation-identity tracing; no independent validation or code execution", "full_empirical_validation": "NO", "provenance": "ND.zip, 2026-10-11; original file retained unchanged" }, "topics": [ "K:Long-running agent execution", "K:Agent failure and escalation", "K:Independent verification and oracles", "K:Evaluation methodology and baselines", "K:Agent context and persistent memory", "K:System understanding and rationale" ], "volumes": [ "V05" ], "book": { "volume": "V05", "title": "Non-Deterministic Loop Engineering", "subtitle": "Design Reliable Agentic AI Systems Beyond the Prompt", "url": "https://www.amazon.com/dp/B0GHCXPBRM", "edition": "1", "source_path": "NDLE/glossary.tex" } } ] }