agentleFS
saker-aisaker-ai/skillhub/skills/SKILL.md

skillhub

Skill7 starsChanged 8 months agoMITView on GitHub

Search, install, publish, and manage skills on a SkillHub registry using curl

  • Sends data out

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. 14790897/my-skills★ 2same name
    2. Vieeeeeee/skillhub★ 2same 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 repositoryOne other file it gives its agents.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 4 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Search by keyword (public, approved, non-deleted only):

    line 53
  2. Very specific:

    Read a file from a version (defaults: version=latest, path=SKILL.md):

    line 100
  3. Distinctive:

    Delete (soft delete, owner/admin only):

    line 145
  4. Distinctive:

    Delete your own rating:

    line 188
  5. Very specific:

    Create a token (scope: full|read|publish):

    line 200
  6. Very specific:

    All endpoints under /api/v1/admin require role=admin.

    line 253
  7. Distinctive:

    Create a token for any user:

    line 283
  8. Very specific:

    When publishing, include a SKILL.md file with YAML frontmatter:

    line 310

≈ “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

SkillHub Operations

Interact with a SkillHub registry to search, install, publish, and manage skills using curl.

Environment Variables

  • SKILLHUB_REGISTRY — Registry URL (default: http://localhost:10070)
  • SKILLHUB_TOKEN — API token for authenticated operations (Bearer token)
  • SKILLHUB_AGENT_SECRET — Shared secret for agent auto-provisioning

Set the base URL:

REGISTRY="${SKILLHUB_REGISTRY:-http://localhost:10070}"

Authenticated requests use Authorization: Bearer $SKILLHUB_TOKEN.

1. Auto-Provisioning (Get Token)

Internal agents can auto-register and obtain an API token using a shared secret:

curl -s -X POST "$REGISTRY/api/v1/agent/provision" \
  -H 'Content-Type: application/json' \
  -d "{\"handle\":\"$(hostname)-agent\",\"secret\":\"$SKILLHUB_AGENT_SECRET\"}" | jq .

Response: {"token":"clh_...","handle":"my-agent","userId":"uuid"}

Save the token for subsequent requests:

TOKEN=$(curl -s -X POST "$REGISTRY/api/v1/agent/provision" \
  -H 'Content-Type: application/json' \
  -d "{\"handle\":\"$(hostname)-agent\",\"secret\":\"$SKILLHUB_AGENT_SECRET\"}" | jq -r .token)
export SKILLHUB_TOKEN="$TOKEN"

The endpoint returns 404 if SKILLHUB_AGENT_SECRET is not set server-side.

2. Search Skills

Search by keyword (public, approved, non-deleted only):

curl -s "$REGISTRY/api/v1/search?q=$ARGUMENTS" | jq '.hits[] | {slug, displayName, summary}'

Query params: q (required), limit (≤100, default 20), offset, sort. Response: {hits: [...], estimatedTotalHits, ...}.

3. List Skills

curl -s "$REGISTRY/api/v1/skills?sort=downloads&limit=20" \
  | jq '.data[] | {slug, displayName, category, downloads}'

Query params: sort (created|updated|downloads|stars, default created), limit (≤100), cursor, category.

Filter by category:

curl -s "$REGISTRY/api/v1/skills?category=devops&sort=downloads" | jq '.data[] | {slug, displayName}'

Paginate with cursor:

NEXT=$(curl -s "$REGISTRY/api/v1/skills?limit=20" | jq -r .nextCursor)
curl -s "$REGISTRY/api/v1/skills?limit=20&cursor=$NEXT" | jq .

Categories: devops, security, data, frontend, backend, infra, testing, ai, general.

4. Inspect a Skill

curl -s "$REGISTRY/api/v1/skills/SLUG" | jq .

List versions:

curl -s "$REGISTRY/api/v1/skills/SLUG/versions" | jq '.versions[] | {version, createdAt}'

Get a specific version:

curl -s "$REGISTRY/api/v1/skills/SLUG/versions/1.0.0" | jq .

Read a file from a version (defaults: version=latest, path=SKILL.md):

curl -s "$REGISTRY/api/v1/skills/SLUG/file?path=SKILL.md&version=latest"

5. Install a Skill

Download as ZIP and extract to a local skills directory:

SLUG="my-skill"
VERSION="latest"
SKILLS_DIR="${HOME}/.skillhub/skills"

mkdir -p "$SKILLS_DIR/$SLUG"
curl -sfL -o /tmp/skill.zip "$REGISTRY/api/v1/download?slug=$SLUG&version=$VERSION"
unzip -o /tmp/skill.zip -d "$SKILLS_DIR/$SLUG"
rm /tmp/skill.zip

Resolve a fingerprint (content hash) to a version:

curl -s "$REGISTRY/api/v1/resolve?fingerprint=HEX" | jq .

6. Publish a Skill

Upload a skill using multipart form. slug and version are required (semver); slug is not auto-inferred from SKILL.md. Requires authentication.

curl -s -X POST "$REGISTRY/api/agent/skills" \
  -H "Authorization: Bearer $SKILLHUB_TOKEN" \
  -F "slug=my-skill" \
  -F "version=1.0.0" \
  -F "displayName=My Skill" \
  -F "files=@./SKILL.md" \
  -F "files=@./script.sh"

Form fields: slug (required), version (required, semver), displayName, visibility, overwrite, files (repeatable; SKILL.md required).

Version must be strictly greater than the current latest. New skills default to visibility=private, moderationStatus=approved.

7. Skill Lifecycle

Delete (soft delete, owner/admin only):

curl -s -X DELETE "$REGISTRY/api/v1/skills/SLUG" -H "Authorization: Bearer $SKILLHUB_TOKEN"

Undelete:

curl -s -X POST "$REGISTRY/api/v1/skills/SLUG/undelete" -H "Authorization: Bearer $SKILLHUB_TOKEN"

Request public visibility (triggers moderation review):

curl -s -X POST "$REGISTRY/api/v1/skills/SLUG/request-public" -H "Authorization: Bearer $SKILLHUB_TOKEN"

Check current identity:

curl -s "$REGISTRY/api/v1/whoami" -H "Authorization: Bearer $SKILLHUB_TOKEN" | jq .

8. Stars

Star / unstar a skill (authenticated):

curl -s -X POST   "$REGISTRY/api/v1/stars/SLUG" -H "Authorization: Bearer $SKILLHUB_TOKEN"
curl -s -X DELETE "$REGISTRY/api/v1/stars/SLUG" -H "Authorization: Bearer $SKILLHUB_TOKEN"

9. Ratings

Rate a skill (score 1-5):

curl -s -X POST "$REGISTRY/api/v1/skills/SLUG/ratings" \
  -H "Authorization: Bearer $SKILLHUB_TOKEN" \
  -H 'Content-Type: application/json' \
  -d '{"score":5,"comment":"great"}'

List ratings (public, cursor paginated):

curl -s "$REGISTRY/api/v1/skills/SLUG/ratings?limit=20" | jq '.data[] | {score, comment, user}'

Delete your own rating:

curl -s -X DELETE "$REGISTRY/api/v1/skills/SLUG/ratings" -H "Authorization: Bearer $SKILLHUB_TOKEN"

10. API Tokens

List your tokens:

curl -s "$REGISTRY/api/v1/tokens" -H "Authorization: Bearer $SKILLHUB_TOKEN" | jq .

Create a token (scope: full|read|publish):

curl -s -X POST "$REGISTRY/api/v1/tokens" \
  -H "Authorization: Bearer $SKILLHUB_TOKEN" \
  -H 'Content-Type: application/json' \
  -d '{"label":"ci-token","scope":"publish","expiresIn":"720h"}'

Revoke a token by ID:

curl -s -X DELETE "$REGISTRY/api/v1/tokens/TOKEN_ID" -H "Authorization: Bearer $SKILLHUB_TOKEN"

11. Namespaces (Organizations)

Create a namespace:

curl -s -X POST "$REGISTRY/api/v1/namespaces" \
  -H "Authorization: Bearer $SKILLHUB_TOKEN" \
  -H 'Content-Type: application/json' \
  -d '{"slug":"acme","displayName":"Acme Inc","description":"...","type":"org"}'

List your namespaces / get / update / delete:

curl -s        "$REGISTRY/api/v1/namespaces"      -H "Authorization: Bearer $SKILLHUB_TOKEN"
curl -s        "$REGISTRY/api/v1/namespaces/acme" -H "Authorization: Bearer $SKILLHUB_TOKEN"
curl -s -X PUT "$REGISTRY/api/v1/namespaces/acme" -H "Authorization: Bearer $SKILLHUB_TOKEN" \
  -H 'Content-Type: application/json' -d '{"displayName":"Acme Corp"}'
curl -s -X DELETE "$REGISTRY/api/v1/namespaces/acme" -H "Authorization: Bearer $SKILLHUB_TOKEN"

Members:

curl -s "$REGISTRY/api/v1/namespaces/acme/members" -H "Authorization: Bearer $SKILLHUB_TOKEN"
curl -s -X POST "$REGISTRY/api/v1/namespaces/acme/members" \
  -H "Authorization: Bearer $SKILLHUB_TOKEN" -H 'Content-Type: application/json' \
  -d '{"handle":"alice","role":"member"}'
curl -s -X DELETE "$REGISTRY/api/v1/namespaces/acme/members/alice" \
  -H "Authorization: Bearer $SKILLHUB_TOKEN"

12. Notifications

curl -s "$REGISTRY/api/v1/notifications?limit=20"       -H "Authorization: Bearer $SKILLHUB_TOKEN" | jq .
curl -s "$REGISTRY/api/v1/notifications/unread"         -H "Authorization: Bearer $SKILLHUB_TOKEN" | jq .unread
curl -s -X POST "$REGISTRY/api/v1/notifications/NOTIF_ID/read" -H "Authorization: Bearer $SKILLHUB_TOKEN"
curl -s -X POST "$REGISTRY/api/v1/notifications/read-all"      -H "Authorization: Bearer $SKILLHUB_TOKEN"

13. Admin Operations

All endpoints under /api/v1/admin require role=admin.

Users:

curl -s "$REGISTRY/api/v1/users"                 -H "Authorization: Bearer $SKILLHUB_TOKEN"
curl -s -X POST "$REGISTRY/api/v1/users"         -H "Authorization: Bearer $SKILLHUB_TOKEN" \
  -H 'Content-Type: application/json' -d '{"handle":"bob","role":"user","email":"b@x.com"}'
curl -s -X POST "$REGISTRY/api/v1/users/role"    -H "Authorization: Bearer $SKILLHUB_TOKEN" \
  -H 'Content-Type: application/json' -d '{"userId":"UUID","role":"moderator"}'
curl -s -X POST "$REGISTRY/api/v1/users/ban"     -H "Authorization: Bearer $SKILLHUB_TOKEN" \
  -H 'Content-Type: application/json' -d '{"userId":"UUID","reason":"spam"}'

Skill moderation:

curl -s "$REGISTRY/api/v1/admin/skills?visibility=pending" -H "Authorization: Bearer $SKILLHUB_TOKEN"
curl -s -X POST "$REGISTRY/api/v1/admin/skills/SLUG/review" \
  -H "Authorization: Bearer $SKILLHUB_TOKEN" -H 'Content-Type: application/json' \
  -d '{"action":"approve"}'      # or "reject"
curl -s -X POST "$REGISTRY/api/v1/admin/skills/SLUG/visibility" \
  -H "Authorization: Bearer $SKILLHUB_TOKEN" -H 'Content-Type: application/json' \
  -d '{"visibility":"public"}'   # or "private"

Audit logs (filter by action, resource_type, actor_id):

curl -s "$REGISTRY/api/v1/admin/audit-logs?limit=50&action=ban_user" \
  -H "Authorization: Bearer $SKILLHUB_TOKEN" | jq .

Create a token for any user:

curl -s -X POST "$REGISTRY/api/v1/admin/tokens" \
  -H "Authorization: Bearer $SKILLHUB_TOKEN" -H 'Content-Type: application/json' \
  -d '{"userId":"UUID","label":"service","scope":"full"}'

14. Webhooks (Git Import)

Incoming webhook endpoints (signature-verified, configured server-side):

POST /api/v1/webhooks/github   # X-Hub-Signature-256
POST /api/v1/webhooks/gitlab   # X-Gitlab-Token
POST /api/v1/webhooks/gitea    # X-Gitea-Signature

These are called by git providers, not by agents. Enable via server config.

15. Health

curl -s "$REGISTRY/healthz"   # liveness
curl -s "$REGISTRY/readyz"    # readiness (DB check)

SKILL.md Format

When publishing, include a SKILL.md file with YAML frontmatter:

---
name: my-skill
description: What this skill does
allowed-tools: Bash
keywords: [example, demo]
---

Skill instructions go here...

Required fields: name, description. Optional: allowed-tools, keywords, user-invocable, argument-hint, arguments, model, when_to_use.

Response Conventions

  • List endpoints return {data: [...], nextCursor: "..."} — paginate by passing cursor back.
  • Search returns {hits: [...], estimatedTotalHits: N}.
  • Mutating endpoints return {message: "..."} or the created resource.
  • Errors return {error: "..."} with appropriate HTTP status (400/401/403/404/500).

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.