mountain-nomad-bc/pacify-x/govern-knowledge-refinery

Claude Code and agent skill by mountain-nomad-bc/pacify-x

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Inventory and admit untrusted knowledge sources, classify semantic novelty, plan non-destructive merges, audit graph coherence, simulate retrieval, and stage approval-bound proposals without writing canon. Use when adding or reconciling knowledge, skills, workflows, formulas, registries, or research corpora.

Install

npx vskill@latest install mountain-nomad-bc/pacify-x/govern-knowledge-refinery

SKILL.md

Govern Knowledge Refinery

  1. Inventory the full source with portable relative paths, byte/hash bounds, rejected symlinks, and redacted secret finding codes. Use the existing non-extracting archive inventory before any archive extraction.
  2. Treat license detection as evidence requiring review, never as an automatic legal decision.
  3. Extract candidates into existing PACIFY-X artifact/task/capability contracts. Record malformed structured inputs separately; do not silently downgrade them to generic text.
  4. Compare identifiers, aliases, structured fields, mechanisms, inputs/outputs, failure modes, invariants, conflicts, negative applicability, evidence, validation, and graph neighborhood.
  5. Assign exactly one of DUPLICATE, ENRICH, VARIANT, CONFLICT, SUPERSEDE, NOVEL, or REVIEW to each candidate. Abstain to REVIEW for close targets or insufficient confidence.
  6. Produce exactly one merge action per decision. Merge plans are proposals and perform no canonical writes.
  7. Audit graph endpoints, duplicate/self edges, provenance state, dependencies, supersession cycles, orphans, and suggested-edge review state.
  8. Reuse canonical capability routing for retrieval simulation, dependency closure, negative matches, authority, safety, budgets, and deterministic tie-breaking.
  9. Calibrate only on separated train and holdout corpora. Reject weight proposals that degrade holdout objectives or forbidden-hit controls.
  10. Stage an approved, hash-bound proposal inside the commissioned project. Canonical apply remains the isolated project control-plane and release-gate responsibility.
  11. Preserve all source, plan, review, graph-diff, test, and rollback evidence. Never hard-delete.

Read runtime contract before admitting, classifying, calibrating, or staging knowledge.

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
9
7-day trend score
17

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/mountain-nomad-bc/pacify-x/govern-knowledge-refinery/badge)](https://verified-skill.com/skills/mountain-nomad-bc/pacify-x/govern-knowledge-refinery)

Something wrong with this skill or its page? Tell us.