agentleFS
thorwhalenthorwhalen/skill/skills/skill-manage/SKILL.md

skill-manage

Skill1 starsChanged 5 months agoMITView on GitHub

Skill management tooling for the skill package. Search, list, inspect, and manage AI agent skills across agents and projects. Use this skill when the user wants to find skills, see what skills are installed, check where skills are linked, inspect a skill's contents, install or uninstall skills, compare skills across agents (Claude Code, Cursor, Copilot), or says things like "what skills do I have", "find a skill for X", "show me the skill", "where are my skills", "install this skill", "remove that skill", "list skills", or "search skills". Trigger on any skill inventory, discovery, or housekeeping request.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. cerebrocybersolutions/libro★ 1same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository5 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 2 are distinctive, and 8 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 8

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Search, list, inspect, and manage AI agent skills.

    line 17
  2. Very specific:

    Read the full contents of an installed skill:

    line 70
  3. Very specific:

    When helping the user understand their skill landscape, check all relevant locations.

    line 178
  4. Very specific:

    Check .claude/skills/ for project-local skills

    line 192
  5. Very specific:

    Check ~/.claude/skills/ for global skills

    line 193
  6. Distinctive:

    Check if it's a symlink pointing to a moved/deleted source

    line 199
  7. Very specific:

    Use the skill-sync skill if the issue is code drift

    line 200
  8. Very specific:

    Cursor and Copilot need format translation: When installing to these targets, the skill is automatically converted — but some metadata is …

    line 207
  9. Very specific:

    Symlinks break if the source moves: If a user reorganizes their project, skill symlinks may dangle.

    line 211
  10. Very specific:

    scope='project' requires being in a project: If find_project_root() can't detect a project (no .git, pyproject.toml, etc.), it will raise …

    line 213

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Skill Manager

Search, list, inspect, and manage AI agent skills.

Core operations

Search for skills

Find skills across local store and remote backends (GitHub, Smithery, Composio, and more):

from skill import search

# Local + remote search
results = search("react best practices", max_results=10)

# Local only (faster, no network)
results = search("testing", local_only=True)

# Specific backends only
results = search("deployment", backends=["github", "smithery"])

CLI:

skill search "react best practices"
skill search "testing" --local-only
skill search "deployment" --max-results 20

Each result is a SkillInfo with:

  • canonical_key — e.g., "thorwhalen/skill-search" (owner/name)
  • name, description — what it does
  • source — where it came from ("local", "github", "smithery", etc.)
  • url — link to the source (for remote results)
  • installed — whether it's already in the local store

List installed skills

from skill import list_skills

skills = list_skills()
for s in skills:
    print(f"{'✓' if s.installed else ' '} {s.canonical_key}: {s.description}")

CLI:

skill list-skills

Inspect a skill

Read the full contents of an installed skill:

from skill import show

s = show("owner/skill-name")
print(s.meta.name)  # 'skill-name'
print(s.meta.description)  # one-liner
print(s.body)  # full instructions
print(s.resources)  # {'scripts': ['run.py'], 'references': ['guide.md']}

CLI:

skill show owner/skill-name

Install a skill

Install from the local store into an agent target:

from skill import install

# Install into project-local Claude Code skills
paths = install("owner/skill-name", scope="project")

# Install globally
paths = install("owner/skill-name", scope="global")

# Install into multiple agents
paths = install("owner/skill-name", agent_targets=["claude-code", "cursor"])

# Copy instead of symlink
paths = install("owner/skill-name", copy=True)

CLI:

skill install owner/skill-name
skill install owner/skill-name --scope global
skill install owner/skill-name --agent-targets claude-code cursor

Uninstall a skill

from skill import uninstall

removed = uninstall("owner/skill-name")

CLI:

skill uninstall owner/skill-name

Link skills from a project

Bulk-link all skills from a project or directory:

from skill import link_skills

# From a project root (auto-discovers .claude/skills/ or {pkg}/data/skills/)
linked = link_skills("/path/to/project")

# Into a specific target
linked = link_skills("/path/to/project", target="~/.claude/skills")

# Force overwrite existing
linked = link_skills("/path/to/project", target=".claude/skills", force=True)

CLI:

skill link-skills /path/to/project
skill link-skills /path/to/project --target ~/.claude/skills --force

Check available search backends

from skill import sources

for src in sources():
    status = "enabled" if src["enabled"] else "disabled"
    print(f"{src['name']}: {status} ({src.get('homepage', 'N/A')})")

CLI:

skill sources

Where skills live

Skills can exist in several places:

LocationPurposeScope
~/.local/share/skill/skills/Local skill store (downloaded/created)User-wide
~/.claude/skills/Claude Code global skillsAll projects
.claude/skills/Claude Code project skillsThis project
.cursor/rules/Cursor rules (translated from SKILL.md)This project
.github/copilot-instructions.mdCopilot instructions (appended)This project
{pkg}/data/skills/Package-shipped skillsDistributed

When helping the user understand their skill landscape, check all relevant locations. Use ls or glob to see what's actually on disk, since symlinks may point to different locations.

Common workflows

"I want to find and use a skill":

  1. search('topic') to find candidates
  2. Review the results — check descriptions and sources
  3. install(key) to link it into the current project
  4. The agent can now use it automatically

"What skills do I have?":

  1. list_skills() for the local store inventory
  2. Check .claude/skills/ for project-local skills
  3. Check ~/.claude/skills/ for global skills
  4. Report which are symlinks vs copies, and where they point

"Something's wrong with a skill":

  1. show(key) to read the full skill
  2. validate(path) to check for structural issues
  3. Check if it's a symlink pointing to a moved/deleted source
  4. Use the skill-sync skill if the issue is code drift

Gotchas

  • install() requires the skill to be in the local store first. If you searched remotely, you need to fetch/create it locally before installing. The create() function can be used to add a skill to the store.
  • Cursor and Copilot need format translation. When installing to these targets, the skill is automatically converted — but some metadata is lost (Cursor doesn't support all frontmatter fields, Copilot appends to a single file).
  • Symlinks break if the source moves. If a user reorganizes their project, skill symlinks may dangle. Check with ls -la and re-link if needed.
  • scope='project' requires being in a project. If find_project_root() can't detect a project (no .git, pyproject.toml, etc.), it will raise RuntimeError. Pass project_dir explicitly in that case.

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.