Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. Start with the skills.sh registry via `npx skills find`; if there are no good matches, fall back to a GitHub deep search for SKILL.md patterns before concluding no skill exists.
.claude/skills/cnfjlhj-find-skills/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | 140% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 147% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 120% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -8% | 0% |
This skill helps you discover and install skills from the open agent skills ecosystem.
npx skills find <query>Notes:
npx skills find with no args).DISABLE_TELEMETRY=1 (or DO_NOT_TRACK=1) when running npx skills ....Use this skill when the user:
The Skills CLI (npx skills) is the package manager for the open agent skills ecosystem. Skills are modular packages that extend agent capabilities with specialized knowledge, workflows, and tools.
Key commands:
npx skills find [query] - Search for skills interactively or by keywordnpx skills add <package> - Install a skill from GitHub or other sourcesnpx skills check - Check for skill updatesnpx skills update - Update all installed skillsBrowse skills at: https://skills.sh/
When a user asks for help with something, identify:
Run the find command with a relevant query:
bashDISABLE_TELEMETRY=1 npx -y skills find [query]
For example:
npx skills find react performancenpx skills find pr reviewnpx skills find changelogThe command will return results like:
Install with npx skills add <owner/repo@skill>
vercel-labs/agent-skills@vercel-react-best-practices
└ https://skills.sh/vercel-labs/agent-skills/vercel-react-best-practicesWhen you find relevant skills, present them to the user with:
Example response:
I found a skill that might help! The "vercel-react-best-practices" skill provides
React and Next.js performance optimization guidelines from Vercel Engineering.
To install it:
npx skills add vercel-labs/agent-skills@vercel-react-best-practices
Learn more: https://skills.sh/vercel-labs/agent-skills/vercel-react-best-practicesIf the user wants to proceed, you can install the skill for them:
bashDISABLE_TELEMETRY=1 npx -y skills add <owner/repo@skill> -g --agent codex -y
The -g flag installs globally (user-level) and -y skips confirmation prompts.
When searching, consider these common categories:
| Category | Example Queries | | --------------- | ---------------------------------------- | | Web Development | react, nextjs, typescript, css, tailwind | | Testing | testing, jest, playwright, e2e | | DevOps | deploy, docker, kubernetes, ci-cd | | Documentation | docs, readme, changelog, api-docs | | Code Quality | review, lint, refactor, best-practices | | Design | ui, ux, design-system, accessibility | | Productivity | workflow, automation, git |
vercel-labs/agent-skills or ComposioHQ/awesome-claude-skillsIf no relevant skills are found in the registry, do a quick GitHub deep search before concluding "no skill exists".
This is useful when a skill exists in a repo but isn't indexed in skills.sh yet.
gh is available, search for skill markdown files by keyword:bash gh search code "KEYWORD path:/^./.?(opencode|ai|llm|claude|codex|agents|pi|cursor|factory)\\/skills?\\/.*\\.md$/" --limit 20
Tips:
playwright e2e, nextjs performance, pr review).SKILL.md and look actively maintained.owner/repo that looks right, list available skills:bash DISABLE_TELEMETRY=1 npx -y skills add owner/repo --list
bash DISABLE_TELEMETRY=1 npx -y skills add owner/repo --skill "<skill-name>" -g --agent codex -y
gh is not available, skip this fallback and:npx skills find ..., orIf there still isn't a good match:
npx skills initExample:
I searched for skills related to "xyz" but didn't find any matches.
I can still help you with this task directly! Would you like me to proceed?
If this is something you do often, you could create your own skill:
npx skills init my-xyz-skill| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | fail→fail | 9,402 | 8,635 | -8% | 1 | 1 | 0% | 1,380 | 1,869 | +35% | 0 | 0 | — |
case-01 | fail→fail | 10,257 | 5,346 | -48% | 1 | 1 | 0% | 1,372 | 1,870 | +36% | 0 | 0 | — |
case-02 | fail→fail | 14,037 | 5,344 | -62% | 1 | 1 | 0% | 1,942 | 1,856 | -4% | 0 | 0 | — |
case-03 | fail→fail | 13,811 | 5,523 | -60% | 1 | 1 | 0% | 1,895 | 1,931 | +2% | 0 | 0 | — |
case-05 | fail→fail | 10,179 | 5,584 | -45% | 1 | 1 | 0% | 1,478 | 1,882 | +27% | 0 | 0 | — |
case-06 | fail→pass | 4,972 | 3,080 | -38% | 1 | 1 | 0% | 891 | 2,134 | +140% | 0 | 0 | — |
case-07 | fail→pass | 5,164 | 3,809 | -26% | 1 | 1 | 0% | 886 | 2,191 | +147% | 0 | 0 | — |
case-08 | fail→pass | 11,597 | 2,959 | -74% | 1 | 1 | 0% | 1,893 | 2,125 | +12% | 0 | 0 | — |
case-09 | fail→pass | 6,846 | 3,622 | -47% | 1 | 1 | 0% | 1,059 | 2,332 | +120% | 0 | 0 | — |
case-10 | fail→fail | 9,189 | 6,151 | -33% | 1 | 1 | 0% | 1,403 | 1,874 | +34% | 0 | 0 | — |
case-11 | fail→fail | 18,224 | 5,215 | -71% | 1 | 1 | 0% | 2,926 | 1,873 | -36% | 0 | 0 | — |
case-12 | fail→fail | 10,809 | 5,051 | -53% | 1 | 1 | 0% | 1,543 | 1,784 | +16% | 0 | 0 | — |
case-13 | fail→fail | 15,911 | 4,984 | -69% | 1 | 1 | 0% | 2,340 | 1,795 | -23% | 0 | 0 | — |
case-14 | fail→fail | 12,640 | 4,542 | -64% | 1 | 1 | 0% | 2,138 | 1,777 | -17% | 0 | 0 | — |
case-15 | fail→fail | 16,834 | 5,423 | -68% | 1 | 1 | 0% | 2,285 | 1,782 | -22% | 0 | 0 | — |
case-16 | fail→pass | 11,761 | 2,087 | -82% | 1 | 1 | 0% | 2,078 | 1,909 | -8% | 0 | 0 | — |
case-17 | fail→pass | 11,789 | 2,793 | -76% | 1 | 1 | 0% | 1,891 | 2,048 | +8% | 0 | 0 | — |
case-18 | fail→pass | 13,451 | 4,920 | -63% | 1 | 1 | 0% | 2,163 | 2,448 | +13% | 0 | 0 | — |
case-19 | fail→pass | 7,273 | 2,761 | -62% | 1 | 1 | 0% | 1,009 | 2,066 | +105% | 0 | 0 | — |
case-20 | pass→pass | 5,936 | 5,917 | -0% | 1 | 1 | 0% | 1,023 | 2,414 | +136% | 0 | 0 | — |
case-21 | pass→pass | 4,595 | 3,497 | -24% | 1 | 1 | 0% | 771 | 2,217 | +188% | 0 | 0 | — |
case-22 | pass→pass | 2,869 | 3,080 | +7% | 1 | 1 | 0% | 410 | 2,084 | +408% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 11 counted toward the lift figure. The other 11 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +36 percentage points is the difference between those two pass rates over the 11 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.