agentleFS
shaunandrewsshaunandrews/agent-skills/skills/agent-skill-inventory-and-consolidation/SKILL.md

agent-skill-inventory-and-consolidation

Skill8 starsChanged 3 months agoMITView on GitHub

Audit skills across agent runtimes, resolve canonical copies, and plan safe consolidation without overwriting divergent repositories.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Verify consolidation after execution by resolving each consumer’s link or registration back to the intended canonical skill and confirming …

    line 24

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “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

Agent Skill Inventory and Consolidation

Procedure

  1. Anchor the audit on every exact skill path or name the requester supplied. Inspect that target before considering similarly named skills from indexes or other runtimes. Complete when each supplied target is mapped to an observed path and identity.
  2. Enumerate each agent runtime’s skill directories, skill configuration entries, workspace-managed skills, and project-local skills. Record whether each entry is a directory, symlink, configured reference, or bundled system/vendor skill. Complete when every runtime has an inventory with source type and location.
  3. Resolve every symlink and compare duplicate names by actual content and target, not by basename alone. Keep similarly named skills separate until their procedures prove equivalent. Complete when duplicates, forks, aliases, and unrelated name collisions are labeled.
  4. Identify canonical ownership for each personal skill: public shared repository, private shared repository, or project repository. Treat agent-local copies as uncategorized until ownership is evidenced. Complete when each personal skill has one proposed canonical class or an explicit unresolved status.
  5. Inspect repository branch, remote, worktree, and deletion state before proposing moves or replacements. Preserve intentional divergence and avoid using a dirty or divergent checkout as a consolidation target without an explicit reconciliation decision. Complete when every proposed canonical repository has a documented safe or blocked state.
  6. Cross-check runtime configuration against the filesystem. Flag configured entries with no target, unconfigured local skills, duplicate registrations, and agent-local copies that would drift from a canonical source. Complete when every discrepancy points to both the configuration entry and filesystem evidence.
  7. Separate the audit from mutation. Produce a change plan that names the canonical source, intended consumers, proposed symlink or registration changes, retirement candidates, and required verification; do not move or delete skills during the audit unless the requester separately approves execution. Complete when every proposed change is individually reviewable and reversible.
  8. Verify consolidation after execution by resolving each consumer’s link or registration back to the intended canonical skill and confirming project-local and vendor/system skills remained in place. Complete when all intended consumers resolve correctly and no protected skill class changed unexpectedly.

Classification Reference

  • Canonical shared skill: maintained once and consumed by multiple agents.
  • Agent-local skill: exists directly under one runtime and may drift.
  • Project-local skill: belongs with one repository and stays there.
  • Vendor/system skill: runtime-managed; exclude from personal consolidation.
  • Alias: different entry that resolves to identical canonical content.
  • Fork: related content with meaningful procedural differences.
  • Name collision: similar name but unrelated purpose; keep distinct.
  • Stale registration: configuration points to no observed skill target.

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.