Open-source repos
awesome-claude-skills (Composio)
Thematic Claude skill curation
This page provides source information and editorial guidance. Recommendations and trust labels do not mean every skill here was tested. Results for individual skills are on the skill data pages. See skill data · Listing policy
Overview
Composio’s awesome-claude-skills is a curated Claude Skills list on GitHub that organizes candidate resources by topic. It is useful for surveying capability areas and then following links to the projects that provide the implementations.
SkillFlux sample check
On 2026-10-31, SkillFlux pinned this source at commit be2a406, drew a seeded sample of 20 from the 64 entries it links to, and ran static health check 1.0.0 on each. 75% had no failing check, said when to use the skill and carried a license.
- 18/20sampled entries had a SKILL.md we could check
- 0/18fail at least one check
- 3/18do not say when to use the skill
- 2/18have no license we could find
- 11/18bundle scripts the agent may run
- 0/18include scripts that call the network
- 2/18include destructive shell commands
- 1/18exceed the 5,000-token body guideline
See the 20 sampled entries
| Entry | Result | Flags |
|---|---|---|
| sanjay3290/ai-skills/skills/outline | 15 pass, 0 warn, 0 fail | Bundles scripts the agent may run |
| anthropics/skills/skills/mcp-builder | 15 pass, 0 warn, 0 fail | Bundles scripts the agent may run |
| jthack/ffuf_claude_skill/ffuf-skill | 12 pass, 3 warn, 0 fail | Bundles scripts the agent may run, No license found |
| anthropics/skills/skills/pdf | 15 pass, 0 warn, 0 fail | Bundles scripts the agent may run |
| NeoLabHQ/context-engineering-kit/plugins/kaizen/skills/kaizen | 14 pass, 1 warn, 0 fail | — |
| yusufkaraaslan/Skill_Seekers/distribution/claude-plugin/skills/skill-builder | 15 pass, 0 warn, 0 fail | — |
| anthropics/skills/skills/xlsx | 15 pass, 0 warn, 0 fail | Bundles scripts the agent may run |
| langchain-ai/langsmith-fetch | No SKILL.md | — |
| BuildShipGrowRepeat/nextjs-sanity-blog-skill/skills/building-blog | 15 pass, 0 warn, 0 fail | — |
| michalparkola/tapestry-skills-for-claude-code/youtube-transcript | 15 pass, 0 warn, 0 fail | — |
| coffeefuelbump/csv-data-summarizer-claude-skill | 11 pass, 4 warn, 0 fail | Bundles scripts the agent may run, No license found |
| obra/superpowers/skills/brainstorming | 13 pass, 2 warn, 0 fail | Scripts contain rm -rf, sudo or similar, Bundles scripts the agent may run |
| rampstackco/claude-skills/dist/pi/.agents/skills/creative-brief-selector | 15 pass, 0 warn, 0 fail | — |
| mhattingpete/claude-skills-marketplace → computer-forensics-skills/skills/computer-forensics | No SKILL.md | — |
| anthropics/skills/skills/pptx | 15 pass, 0 warn, 0 fail | Bundles scripts the agent may run |
| conorluddy/ios-simulator-skill/ios-simulator-skill/skills/ios-simulator-skill | 13 pass, 2 warn, 0 fail | Scripts contain rm -rf, sudo or similar, Bundles scripts the agent may run |
| obra/superpowers/skills/finishing-a-development-branch | 15 pass, 0 warn, 0 fail | — |
| uxKero/anydesign | 15 pass, 0 warn, 0 fail | Bundles scripts the agent may run |
| omkamal/pypict-claude-skill | 12 pass, 3 warn, 0 fail | Bundles scripts the agent may run |
| NeoLabHQ/context-engineering-kit/plugins/sadd/skills/subagent-driven-development | 15 pass, 0 warn, 0 fail | — |
The sample reads files only; it does not run any skill and does not describe entries outside the sample. Flags describe what a skill contains; they are not verdicts. Raw data · Method
Best for
- Users surveying Claude skill areas before inspecting individual resources.
- Teams building a shortlist for writing, media or development work.
Use cases
- Find the right skill
- Content & writing
- Video & media
SkillFlux notes
Read the awesome list as an index and distinguish its curator from each skill’s maintainer. Once you choose an entry, inspect the source project. The list’s general description is not a capability or support commitment for every linked item.
Installation and use
Open the original repository and read its README, the selected skill instructions and any linked scripts. Check installation paths, host requirements, dependencies and licenses, then follow the project documentation to verify a small task in a workspace you can restore.
Limitations
- Inclusion does not establish identical execution testing or licensing across linked projects.
Alternatives
- awesome-agent-skills (VoltAgent)
Compare VoltAgent’s list for discovery across vendor and community sources in the broader agent ecosystem.
- awesome-claude-skills (travisvn)
Read travisvn’s list for another Claude-focused perspective alongside architecture explanations.