agentleFS
loganware05loganware05/captains-compass-cursor/.cursor/skills/skill-lifecycle/SKILL.md

skill-lifecycle

Skill2 starsChanged 32 days agoMITView on GitHub

Advances candidates through APPROVED/AVAILABLE/PROVEN and tracks Skill proficiency via Experiences

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    After the Skill is used successfully in real workstreams, record Experiences (execution-telemetry / record-execution-run.sh) with …

    line 54
  2. Very specific:

    Every Skill has an immutable inode at .cursor/skills/inodes/<sha256>.json (SHA-256 over SKILL.md + capability.yaml).

    line 75

“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Skill Lifecycle

Use this Skill when

A Technology Intelligence candidate has reached SANDBOX_TESTED and the Captain wants to advance it through:

APPROVED → AVAILABLE_SKILL → PROVEN_SKILL

or an existing Skill should graduate toward PROVEN proficiency using successful Experience evidence (training loop).

For earlier stages (DISCOVERED → SANDBOX_TESTED) use Skill candidate-promotion.

Inputs

  • Candidate JSON (staging or fixture) at SANDBOX_TESTED or later
  • Evidence paths
  • Explicit --captain-approved for every post-sandbox stage
  • Skill slug (for AVAILABLE proposals and Experience matching)
  • Experience store under .agent/experience/ (for PROVEN)

Procedure

  1. Confirm candidate is at least SANDBOX_TESTED and approved_for_execution is false.
  2. Captain-approved advance to APPROVED:

    ./scripts/promote-candidate.sh --candidate <path.json> \
      --stage APPROVED \
      --evidence .agent/evidence/approval.md \
      --captain-approved \
      --skill-slug <slug>
  3. Emit AVAILABLE_SKILL install proposal (never live install):

    ./scripts/promote-candidate.sh --candidate <staging.json> \
      --stage AVAILABLE_SKILL \
      --evidence .agent/evidence/approval.md \
      --captain-approved \
      --skill-slug <slug>

    Review .agent/capabilities/candidates/available-proposals/<slug>/. Open a Captain-reviewed PR to copy into .cursor/skills/<slug>/ only after approval.

  4. After the Skill is used successfully in real workstreams, record Experiences (execution-telemetry / record-execution-run.sh) with skills_used including the slug. When ≥2 successful Experiences exist, advance to PROVEN_SKILL:

    ./scripts/promote-candidate.sh --candidate <staging.json> \
      --stage PROVEN_SKILL \
      --evidence .agent/evidence/proven.md \
      --captain-approved \
      --skill-slug <slug>

    Override threshold with COMPASS_PROVEN_SUCCESS_THRESHOLD if needed.

  5. Optional training draft from Experience: Skill experience-skill-training (./scripts/train-skill-from-experience.sh). Learning-run Experiences: Skill skill-learning-loop (bridge-learning-experiences.sh).
  6. Re-run ./scripts/compile-capability-registry.sh after any live Skill PR merges.

M40: content-addressed Skill inodes

Every Skill has an immutable inode at .cursor/skills/inodes/<sha256>.json (SHA-256 over SKILL.md + capability.yaml). Skill identity is content, not slug:

  • Editing a Skill produces a new inode; the old inode is immutable history.
  • Experience/proficiency reputation pins to the inode hash it was earned on.
  • Reputation carry-over to a changed Skill's new inode requires ./scripts/build-skill-inodes.sh --captain-approved (Captain gate); otherwise the new inode carries reputation.captain_approved: false.
  • doctor.sh fails closed when committed inodes are stale — rerun ./scripts/build-skill-inodes.sh after editing any Skill.
  • The compiled registry embeds provenance.content_hash + skill_inode per Skill entry.

Output

  • Staging candidate with updated lifecycle_stage
  • AVAILABLE proposal under .agent/capabilities/candidates/available-proposals/
  • PROVEN staging record when Experience threshold met

Prohibited actions

  • Auto-installing into .cursor/skills/
  • Setting approved_for_execution: true
  • Advancing past SANDBOX_TESTED without --captain-approved
  • Mutating matcher weights from lifecycle advances

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.