DATA block to be filled by the build script (all ready skills, full skill.md content).
Everything below will light up once the data lands.
239 SOPs — every task documented to the definitive-article standard, so humans and AI agents run them the same way.
This library is not a wiki that rots. It is a loop that runs. Every task executed by the team — human or AI — passes through five gates, and the documentation comes out sharper than it went in.
Do → Document (skill.md) → QA (Definition of Done) → Example (Meta-Article) → Improve (SOP Update) → ↺
Every task we perform is captured as a skill.md — a step-by-step SOP with a QA gate (the Definition of Done) and a worked example (the meta-article). The next person to run the task, human or AI agent, starts from the best documented run instead of from scratch.
Each pass through the loop sharpens the procedure, tightens the checklist, and adds proof. Nothing is ever done twice from zero. The system compounds.
239 tasks documented to the definitive-article standard, generated and QA'd by parallel AI workers (slug/section/URL/status checks all passed).
Every future run appends an entry here. The changelog is the proof of recursion.
Nothing in 239 SOPs matches that combination of search and status filter.