Nümtema MCP Foundry
Transforms OpenAPI specs into governed MCP applications with a local-first studio, OAuth, simulation, and Docker deployment.
README
Nümtema MCP Foundry v1.3.0
Transformez une API OpenAPI en application MCP gouvernée, testable, connectable à ChatGPT et désormais pilotable depuis un Studio local-first.
OpenAPI
→ Foundry Studio
→ inspection et CapabilityMap
→ édition gouvernée des tools
→ OAuth / provider / approvals
→ simulation sans réseau
→ runtime MCP + ChatGPT App
→ package Docker Coolify/VPS
Installation
Node.js 22 ou supérieur :
npm install --global ./release/numtema-mcp-foundry-1.3.0.tgz
foundry doctor
Depuis le ZIP :
./scripts/install.sh
Windows PowerShell :
.\scripts\install.ps1
Lancer le Studio
foundry studio init ./mon-foundry-studio --name "Mon MCP"
foundry studio serve ./mon-foundry-studio
Ouvrez ensuite l’URL locale affichée par la commande. Le Studio comprend :
- import OpenAPI JSON/YAML ;
- inspection et classification des risques ;
- édition des tools et scopes ;
- configuration provider et OAuth ;
- aperçu du widget d’approbation ;
- simulation HTTP dry-run ;
- génération d’un package Docker/Coolify/VPS autonome.
Pour générer sans interface :
foundry studio build ./mon-foundry-studio
Le package est créé dans :
mon-foundry-studio/deploy/package
Runtime ChatGPT App
La v1.3 conserve le gateway v1.1 : OAuth 2.1 + PKCE, resources Apps SDK, widget sécurisé, scopes progressifs, Ledger durable et reçus Ed25519.
foundry app init ./numtema-chatgpt-app \
--public-base-url https://mcp.example.com
Commandes principales
foundry doctor / demo / init
foundry inspect / map / compile / validate
foundry adapters / plan
foundry auth-bindings / auth-plan / preflight
foundry ledger / dispatch
foundry mcp init / inspect / smoke / serve
foundry app init / inspect / smoke / serve
foundry studio init / inspect / build / serve
Sécurité
- aucune valeur de credential dans les plans, traces ou reçus ;
- credentials injectés uniquement à la frontière HTTP ;
- risk/approval downgrade refusé par le Studio ;
- serveur Studio limité au loopback et écritures protégées par CSRF ;
- package de déploiement sans clé privée ;
- clés Ed25519 générées au premier démarrage dans un volume durable ;
- nonce, idempotence et budget réservés dans le Ledger ;
- zéro dépendance runtime.
GitHub and Coolify
- Agent instructions:
AGENTS.md - GitHub publication:
docs/deployment/GITHUB_PUBLISHING.md - Exact Coolify settings:
docs/deployment/COOLIFY_CONFIGURATION.md
Before publishing:
./scripts/audit-secrets.sh
npm run typecheck
npm test
Documentation
Frontière v1.3
La v1.3 produit un package prêt à importer dans Coolify ou à lancer sur un VPS. Elle ne se connecte pas encore directement à un compte Coolify et ne déclenche pas un déploiement distant avec votre token API.
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
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.