# agentskill.work

agentskill.work is a curated, searchable index of trending Claude Skill GitHub repositories.

More details for agents/LLMs:
- https://agentskill.work/llms-full.txt

## Canonical host / duplicates
- Canonical host: https://agentskill.work
- https://www.agentskill.work redirects to https://agentskill.work

## Purpose
- Help users discover active Claude Skill projects quickly
- Provide concise repo summaries with core stats and topics

## Content model (per skill)
- full_name (owner/repo)
- description (original)
- description_zh (machine translation when enabled)
- summary_en / summary_zh (offline LLM enrichment)
- key_features_en / key_features_zh (offline LLM enrichment)
- use_cases_en / use_cases_zh (offline LLM enrichment)
- seo_title_en / seo_title_zh, seo_description_en / seo_description_zh (offline LLM enrichment)
- stars, forks, language
- topics (stored as comma-separated string)
- last_pushed_at (GitHub), fetched_at (last sync)
- repo_created_at, repo_updated_at (GitHub)
- html_url, repo_id

## Pages
- Home list (zh): https://agentskill.work/zh
- Home list (en): https://agentskill.work/en
- Latest list: https://agentskill.work/{lang}/latest
- Detail page: https://agentskill.work/{lang}/skills/{owner}/{repo}
- Facets (SEO entry pages):
  - Topics: https://agentskill.work/{lang}/topics/{topic}
  - Languages: https://agentskill.work/{lang}/languages/{language}
  - Owners: https://agentskill.work/{lang}/owners/{owner}

## Language
- zh and en are supported
- "Claude Skill" is kept as a proper term

## Data sources and freshness
- Source: GitHub Search API
- Sync: scheduled background job (typically hourly)
- Sync uses two strategies: trending (stars) + latest discovery (created window, sort=updated)
- Sync may stop early to respect GitHub rate limits
- Search keywords include: "Claude Skill" and "agent skill"

## SEO and structured data
- WebSite, FAQ, and ItemList on the home page
- SoftwareSourceCode and BreadcrumbList on detail pages
- Sitemap index: https://agentskill.work/sitemap-index.xml
- Sitemap alias: https://agentskill.work/sitemap.xml (same as sitemap index)
- Robots: https://agentskill.work/robots.txt

## API (public)
- Base path: https://agentskill.work/api
- OpenAPI: https://agentskill.work/api/openapi.json
- List: /skills
  - query: `q`, `topic`, `language`, `owner`, `sort`, `limit`, `offset`
  - sort: `stars` (default) | `newest`
- Detail: /skills/{owner}/{repo}
- Facets: /facets/topics, /facets/languages, /facets/owners
- Please avoid large-scale crawling; respect rate limits

## Data usage and attribution
- Metadata is aggregated from GitHub; the source of truth is each repository
- Follow GitHub’s terms and each repo’s own license
- When citing a project, link to the GitHub repo and/or the detail page
- Do not imply endorsement by agentskill.work or the repository owners

## Crawl guidance
- Prefer canonical URLs without tracking parameters
- Use detail pages for attribution when citing a repository
- Pagination for list pages is crawlable via `?offset=...` URLs (no need to execute JS)

## Contact
- https://agentskill.work/