English Français العربية

Skill History

Registre officiel Recherche et web

Track download history for 70,000+ agent skills. Search and get daily snapshots.

Code source Documentation

Installation

Ajoutez ceci à la configuration MCP de votre client (Claude Desktop : claude_desktop_config.json ; Cursor : .cursor/mcp.json ; VS Code et les autres utilisent la même forme « mcpServers »).

Remote (streamable-http)
Claude Code 
claude mcp add --transport http skill-history 'https://skill-history.com/mcp'
Configuration JSON (Claude Desktop, Cursor, VS Code) 
{
    "mcpServers": {
        "skill-history": {
            "type": "http",
            "url": "https://skill-history.com/mcp"
        }
    }
}

Documentation

Tirée du README du projet sur GitHub, actualisée chaque semaine. Les liens ouvrent GitHub.

Sommaire
  1. skill-history
  2. How it works
  3. Architecture
  4. API
  5. Local development
  6. Roadmap
  7. Contributing
  8. License

skill-history

Track and visualize agent skill download history from ClawHub. An embeddable chart for READMEs — star-history, but for agent skills.

skill-history.com · built by Pineapple AI

How it works

skill-history.com tracks daily download counts for all 54,000+ skills on ClawHub. Every skill gets a chart page and an embeddable SVG:

  • Chart page — skill-history.com/{handle}/{slug}
  • SVG embed — skill-history.com/chart/{handle}/{slug}.svg
  • Compact badge — skill-history.com/badge/{handle}/{slug}.svg

Your {handle} is your GitHub username (ClawHub uses GitHub OAuth), and {slug} is your skill's slug on ClawHub.

Add to your README
[![Download History](https://skill-history.com/chart/YOUR_HANDLE/YOUR_SLUG.svg)](https://skill-history.com/YOUR_HANDLE/YOUR_SLUG)

Or use the link generator at skill-history.com.

Architecture

  • Cloudflare Workers + Hono — serves the site, chart SVGs, and scheduled sweeps
  • Cloudflare D1 (SQLite) — stores skills and daily snapshots
  • Edge cache — GET responses are stored in Cloudflare's cache via the Cache API (caches.default), so D1 only sees cache misses; weekly blog data is materialised in D1 once per week
  • d3-shape — smooth monotone curves (same algorithm as star-history)
  • Data source — ClawHub public API, sweeps every 2 hours

API

All skill pages return JSON when requested with Accept: application/json:

curl -H "Accept: application/json" https://skill-history.com/gavinlinasd/self-preserve

Full OpenAPI spec at skill-history.com/api/openapi.json.

For AI agents: skill-history.com/llms.txt.

Local development

npm install

# Authenticate with Cloudflare
wrangler login

# Create D1 database, copy database_id into wrangler.jsonc
wrangler d1 create skill-history

# Apply schema to local dev DB
npm run db:migrate:local

# Run locally
npm run dev

Roadmap

  • [x] Daily download tracking for all ClawHub skills
  • [x] Embeddable SVG chart + compact badge
  • [x] Landing page with link generator
  • [x] Agent discoverability (llms.txt, OpenAPI)
  • [ ] Smooth d3-powered chart rendering (full axis upgrade)
  • [ ] Multi-source tracking (GitHub stars, non-ClawHub installs)
  • [ ] Comparison charts (overlay multiple skills)
  • [ ] Weekly "fastest growing skills" reports
  • [ ] MCP server for agent tool access

Have an idea? Open an issue or submit a PR — we'd love your input.

Contributing

This is a brand new project and we're actively shaping it based on what skill authors actually want. Contributions welcome:

  • Feature requests — open an issue describing what you'd like to see
  • Bug reports — if a chart looks wrong or a skill is missing, let us know
  • Pull requests — code, docs, or design improvements are all welcome
  • Data sources — ideas for tracking skills beyond ClawHub (GitHub stars, other registries)

License

MIT

Questions fréquentes

Qu'est-ce que Skill History ?

Track download history for 70,000+ agent skills. Search and get daily snapshots.

Comment installer Skill History ?

Il fonctionne comme serveur distant. Dans Claude Code : claude mcp add --transport http skill-history 'https://skill-history.com/mcp'. Pour les autres clients, collez la configuration JSON de cette page dans leurs réglages MCP.

Faut-il une clé API pour Skill History ?

Son entrée au registre n'indique aucune clé ni secret obligatoire. Certaines fonctions peuvent tout de même demander un compte auprès du service auquel il se connecte.

Skill History est-il open source ?

Oui. Son code est sur GitHub sous licence MIT.

Qui maintient Skill History ?

Il est publié par pineapple-farm. AI.tn le référence sans le maintenir ni le recommander.

Issu du registre MCP officiel. Être listé n'est pas une recommandation : vérifiez le code et les permissions d'un serveur avant de lui donner accès à vos données.