A living, queryable knowledge graph on how the full software product development lifecycle becomes AI-native (AI-DLC / AIDLC / Agentic SDLC) — across technology, process, governance, org design, and roles.
Built to be read two ways: by an agent (e.g. Claude Code, reasoning over structured frontmatter) and by a human (Obsidian graph + Dataview). Every claim is decision-grade — cited, dated, and rated for how fast it goes stale.
Start here → Home.md
phases/— the 12-phase lifecycle spine (discovery → support), deliberately beyond coding & testingconcepts/— cross-cutting concepts & the AI-native foundation (context layer, learning loop, …)cases/— credibility-flagged real-world case evidence (success & cautionary)signals/— monitored industry watchlists (e.g. JD signals), tracked over timemeta/+_*.md— schema, templates, saved queries, freshness policy, operating guidesskills/— thekg-*instruction-skills that keep the vault current
- Read — open
Home.mdin Obsidian (or any Markdown viewer). - Query — filter on frontmatter fields; contract in
_Schema.md, saved views in_Queries.md. - Maintain — the
kg-*skills; when to use which + recipes inmeta/Using-the-Skills.md. - Trust — each node carries
volatility+next_review; freshness policy in_Freshness.md.
Every node conforms to the controlled-vocabulary schema in _Schema.md. Nodes are interlinked with
Obsidian wiki-links (the [[Node]] syntax) — these resolve in Obsidian or any wikilink-aware tool, not plain Markdown.
Built June 2026. Internal knowledge base; owners TBD pending assignment.