mcp-moncompte
Read-only MCP server exposing the catalog of articles from moncompte.org. Enables AI agents to search and retrieve article content via tools.
README
mcp-moncompte
Serveur MCP (Model Context Protocol) en lecture seule qui expose le catalogue de notices de moncompte.org aux agents IA (Claude, etc.), déployé sur Cloudflare Workers.
- Stack : TypeScript + Cloudflare Workers + Durable Objects + SDK MCP officiel + Zod
- Source de données : endpoints publics du site statique moncompte.org (Astro / Cloudflare
Pages) — pas de WP REST. La recherche et
get_articlelisent/search-index.json(index plein-texte généré au build : titre, slug, description, corps complet). La ressourceaboutproxifie/llms.txt. Aucun secret requis. - Cache : edge Cloudflare 1 h sur chaque appel
Dépendance :
search_articlesetget_articlereposent surhttps://moncompte.org/search-index.json, produit par le site Astro (endpointsrc/pages/search-index.json.js). Cet index exclut déjà les articles datés dans le futur (isPublished) et les brouillons (draft) — le MCP ne voit donc que le contenu réellement en ligne. L'index est régénéré à chaque build/déploiement du site.
Outils & ressource exposés
| Type | Nom | Rôle |
|---|---|---|
| Outil | search_articles |
Recherche par mots-clés (titre, slug, URL, extrait) |
| Outil | get_article |
Contenu complet d'un article (par slug) |
| Ressource | moncompte://about |
Proxy du llms.txt de moncompte.org |
Endpoints
/sse— Server-Sent Events (clients historiques)/mcp— Streamable HTTP (transport recommandé)- Rate-limit : 100 req / 60 s par IP (binding natif Workers)
- Durable Object :
NoticielMCP(classeMcpAgent)
Développement
npm install
npm run dev # wrangler dev (local)
npm run typecheck # tsc --noEmit
Déploiement
npx wrangler login # une fois (compte Cloudflare)
npm run deploy # wrangler deploy
Le binding de rate-limit (
[[unsafe.bindings]]typeratelimit) requiert Wrangler ≥ 4.36. Le Durable Object McpAgent utilise le stockage SQLite (migrationnew_sqlite_classes).
Connecter un client MCP
- URL :
https://mcp-moncompte.<ton-sous-domaine>.workers.dev/mcp(ou/sse) - Claude Desktop / autres : via
mcp-remote:
{
"mcpServers": {
"moncompte": {
"command": "npx",
"args": ["mcp-remote", "https://mcp-moncompte.<sous-domaine>.workers.dev/sse"]
}
}
}
Notes
- Aucune écriture : le serveur ne fait que lire l'API publique de moncompte.org.
- Les versions de
agentset@modelcontextprotocol/sdkévoluent vite : en cas d'erreur de build, fairenpm install agents@latest @modelcontextprotocol/sdk@latest.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.