operationalizing-expertise
Made by JeffOperationalize expert methods into corpus, quote bank, triangulated kernel, operator library, and validators. Use when distilling a methodology or mining session history into executable rules.
Works With
Quality and Utility come from canonical skill metadata. Community ratings appear separately as stars when subscriber reviews exist.
Subscribe to download
Public previews are available, but downloads, CLI sync, and premium content require an active subscription.
Sign In to Subscribe- Full skill content with deterministic versioning
- CLI sync + local search index
- Access to Jeffrey’s premium library
Save to your account
Saved skills live in your cloud list. Run jsm sync on each machine to pull them locally.
What's New with this Skill
1Updates, enhancements, and release notes for this skill.
New: Operationalizing Expertise, turn a methodology into corpus, kernel, operators, and validators
Expertise lives in the head. To use it across many sessions, projects, and agents, you have to externalize it into auditable artifacts. Three tracks: static distillation (corpus → quote bank → triangulated kernel → operator library → validators), orchestrated loops (kickoff → deltas → deterministic merge + lint), and session mining (your repeated behaviors from cass history become "when X, do Y because Z" rules). Validation scripts block release when they fail.
- Three tracks for three starting points: A (static distillation from transcripts/papers), B (orchestrated multi-agent sessions), C (session mining from cass history)
- Triangulation across three model distillations (GPT, Opus, Gemini): only consensus goes into the kernel, disputed items go to an appendix, and items a single model found are discarded
- Evidence-first: every rule cites anchors in the quote bank, and provenance (quote ID, source file, page or section) must be auditable
- Operator cards carry a definition, at least 3 when-to-use triggers, failure modes, a copy-paste prompt module, a canonical tag and quote-bank anchors. The library also names standard composition chains
- Validators are scripts, not advice: validate-corpus.py, validate-operators.py and extract-kernel.py check anchors, quote-bank entries, kernel markers and required operator fields. A failing gate blocks release
- Mandatory artifacts: corpus/primary_sources/, corpus/quote_bank/quote_bank.md, corpus/specs/triangulated_kernel.md, corpus/specs/operator_library.md, kickoff templates, and the validation scripts
- The brenner_bot case study walks through a complete Track B implementation, from the 236-segment transcript to the kernel and kickoff messages
Track A: Static Distillation. Use when you have transcripts or papers and want a stable, parseable methodology (Lemelson-style). The flow is corpus → quote bank → multi-model distillation → triangulated kernel → operator library → validators. Triangulation runs three models (GPT, Opus, Gemini) on the same sources and keeps only what all three agree on: consensus goes into the kernel, disputed items go to an appendix, and items only one model found are discarded.