agentleFS
dees91dees91/agent-skill-manager/CLAUDE.md

agent-skill-manager

CLAUDE.md1 starsChanged 3 months agoMITView on GitHub

This repository uses AGENTS.md as the canonical project brief for all agents. Before making code or documentation changes: After non-trivial work that creates reusable knowledge, update the relevant wiki topic/source pages and append to docs/wiki/log.md. The wiki is a synthesis layer; verify current behavior against code and tests and keep accepted product decisions in AGENTS.md and the relevant planning file. Do not implement behavior that contradicts the decisions recorded in AGENTS.md without first updating the plan with the user.

The file

Show its rules and text

We split this CLAUDE.md into the rules it gives an agent

21 rules. About 0 are common, 20 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈20

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before making code or documentation changes:

    line 5
  2. Distinctive:

    Read docs/wiki/README.md and docs/wiki/index.md, then open the topic pages relevant to the task.

    line 8
  3. Distinctive:

    If working on the MVP implementation, read planning/phase-1-mvp-tasks.md.

    line 9
  4. Distinctive:

    If working on group or bulk toggle behavior, read planning/phase-2-group-bulk-toggle-tasks.md.

    line 10
  5. Distinctive:

    If working on repository installation behavior or documentation, read planning/phase-3-repo-install-tasks.md.

    line 11
  6. Distinctive:

    If working on repository update or uninstall behavior or documentation, read planning/phase-4-repo-update-uninstall-tasks.md.

    line 12
  7. Distinctive:

    If working on local path install or uninstall behavior or documentation, read planning/phase-5-local-path-install-tasks.md.

    line 13
  8. Distinctive:

    If working on the desktop app, read the relevant Phase 6-10 planning file.

    line 14
  9. Distinctive:

    If working on publication or repository hygiene, read planning/phase-11-publication-readiness-tasks.md.

    line 15
  10. Distinctive:

    If working on GitHub release artifacts, read planning/phase-12-github-release-tasks.md.

    line 16
  11. Distinctive:

    If working on managed checkout diagnosis or repair, read planning/phase-21-checkout-repair-tasks.md.

    line 17
  12. Distinctive:

    If working on reporting or installing skills added to a managed repository after install, read …

    line 18
  13. Distinctive:

    If working on installing skills directly as OFF, read planning/phase-23-install-as-off-tasks.md.

    line 19
  14. Distinctive:

    If working on optional TypeSafe recommendations or BYOK settings, read planning/phase-24-typesafe-advisor-tasks.md.

    line 20
  15. Distinctive:

    If working on duplicate skill names inside one install source, read planning/phase-25-duplicate-skill-discovery-tasks.md.

    line 21
  16. Very specific:

    If working on installing a skill under another name when another source owns its name (--as), read …

    line 22
  17. Distinctive:

    If working on removing some recorded skills from a source, or on the update failure when a repository removed an installed skill, read …

    line 23
  18. Distinctive:

    If working on the desktop advisor receipts view or on forgetting a blocked advisor receipt, read …

    line 24
  19. Distinctive:

    Follow the task status rules in the relevant planning file before starting and after finishing work.

    line 25
  20. Distinctive:

    After non-trivial work that creates reusable knowledge, update the relevant wiki topic/source pages and append to docs/wiki/log.md.

    line 27
  21. Distinctive:

    Do not implement behavior that contradicts the decisions recorded in AGENTS.md without first updating the plan with the user.

    line 29

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md 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
CLAUDE.md

Claude Code Guidance

This repository uses AGENTS.md as the canonical project brief for all agents.

Before making code or documentation changes:

  1. Read AGENTS.md.
  2. Read docs/wiki/README.md and docs/wiki/index.md, then open the topic pages relevant to the task.
  3. If working on the MVP implementation, read planning/phase-1-mvp-tasks.md.
  4. If working on group or bulk toggle behavior, read planning/phase-2-group-bulk-toggle-tasks.md.
  5. If working on repository installation behavior or documentation, read planning/phase-3-repo-install-tasks.md.
  6. If working on repository update or uninstall behavior or documentation, read planning/phase-4-repo-update-uninstall-tasks.md.
  7. If working on local path install or uninstall behavior or documentation, read planning/phase-5-local-path-install-tasks.md.
  8. If working on the desktop app, read the relevant Phase 6-10 planning file.
  9. If working on publication or repository hygiene, read planning/phase-11-publication-readiness-tasks.md.
  10. If working on GitHub release artifacts, read planning/phase-12-github-release-tasks.md.
  11. If working on managed checkout diagnosis or repair, read planning/phase-21-checkout-repair-tasks.md.
  12. If working on reporting or installing skills added to a managed repository after install, read planning/phase-22-new-skill-discovery-tasks.md.
  13. If working on installing skills directly as OFF, read planning/phase-23-install-as-off-tasks.md.
  14. If working on optional TypeSafe recommendations or BYOK settings, read planning/phase-24-typesafe-advisor-tasks.md.
  15. If working on duplicate skill names inside one install source, read planning/phase-25-duplicate-skill-discovery-tasks.md.
  16. If working on installing a skill under another name when another source owns its name (--as), read planning/phase-26-cross-source-skill-alias-tasks.md.
  17. If working on removing some recorded skills from a source, or on the update failure when a repository removed an installed skill, read planning/phase-27-source-skill-removal-tasks.md.
  18. If working on the desktop advisor receipts view or on forgetting a blocked advisor receipt, read planning/phase-28-advisor-receipts-view-tasks.md.
  19. Follow the task status rules in the relevant planning file before starting and after finishing work.

After non-trivial work that creates reusable knowledge, update the relevant wiki topic/source pages and append to docs/wiki/log.md. The wiki is a synthesis layer; verify current behavior against code and tests and keep accepted product decisions in AGENTS.md and the relevant planning file.

Do not implement behavior that contradicts the decisions recorded in AGENTS.md without first updating the plan with the user.

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.