MobileBuildMCP
An MCP server and CLI that lets AI agents build, test and run iOS and macOS projects.
n8n-mcp est un projet de agents IA open source, créé par czlonkowski (czlonkowski/n8n-mcp sur GitHub). Il compte 23 049 étoiles sur GitHub et 3 663 forks, il est publié sous la licence MIT et a été mis à jour pour la dernière fois sur GitHub le 6 octobre 2026.
An MCP server that lets Claude, Cursor and Windsurf build n8n workflows for you
Try the hosted dashboard for the fastest start, or self-host the server next to your own n8n instance.
Clone the repository, or use the hosted version with an API key from the project dashboard.
Add the n8n-MCP server to Claude Desktop, Claude Code, Cursor or Windsurf, with your n8n instance details.
Ask your assistant to design a workflow, validate each node, and create it in n8n.
Gives your AI assistant accurate documentation and configuration details for n8n nodes.
Validates nodes (minimal and full modes) and whole workflows before they are deployed, catching bad configuration early.
Can return real configurations from workflow templates so the assistant starts from working examples.
Through your n8n instance, the assistant can create and manage workflows, executions and related resources.
Designed for Claude Desktop, Claude Code, Windsurf and Cursor.
Use the hosted dashboard (free tier included) or self-host with npx, Docker or a local install per the README's guide.
Get the source code from GitHub and enter the project directory.
git clone https://github.com/czlonkowski/n8n-mcp.git
cd n8n-mcpPoint your MCP client at the self-hosted server (the README's self-hosting guide covers npx, Docker and local options).
{
"mcpServers": {
"n8n-mcp": {
"command": "npx",
"args": ["-y", "n8n-mcp"]
}
}
}
# see README self-hosting guide for the exact optionsRestart your MCP client, then ask the assistant to validate a node or draft a small workflow.
# Restart Claude Desktop / Cursor, then prompt:
# Build an n8n workflow that posts new RSS items to SlackIt is maintained by Romuald Czlonkowski and contributors on GitHub.
Yes - the server is MIT licensed and free. The optional hosted dashboard has a free tier (100 tool calls/day per the README) with paid tiers beyond that.
You need access to an n8n instance - its API key - for the assistant to create real workflows, and your AI client's own subscription or API key. n8n itself can be self-hosted for free.
Compared with asking a chatbot to write n8n JSON from memory, n8n-MCP validates nodes and workflows against real n8n schemas and templates, which sharply reduces broken imports.
Clone the repo or sign up for the hosted dashboard, add the server to your MCP client config, restart the client, and ask it to build a small test workflow.
An MCP server and CLI that lets AI agents build, test and run iOS and macOS projects.
JADX plugin + MCP server that lets LLMs analyze Android APKs live — decompile, review, find vulnerabilities
Production-grade MCP server for Godot — let AI build scenes, edit scripts and wire your game live
Chrome DevTools for coding agents: an MCP server that lets agents debug and inspect real Chrome
MCP server that lets AI agents run thousands of Apify Store scrapers and crawlers as tools.
Connect your Zotero research library to Claude and other AI assistants through MCP.
Les autres projets de la catégorie Agents IA avec le plus d'étoiles du catalogue LibreAI — la sélection réaliste d'alternatives si n8n-mcp ne vous convient pas.
A personal AI that really does things — on any OS, with your data staying yours.
An agentic skills framework & software development methodology that works.
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
The agent that grows with you — a Python framework for adaptive AI agents.
DeepSeek's plugin-based agent harness where everything is a plugin.
Projets de la catégorie Agents IA proches dans le catalogue, les plus proches en étoiles de n8n-mcp — continuez à explorer de part et d'autre.
A framework for building agents and services
NVIDIA's reference stack for running AI agents more securely inside OpenShell sandboxes
One RAG framework for everything — including multimodal documents.
Open-source LLM observability: trace, evaluate and monitor apps and agents
Jev Ultrafast — the fastest browser agent from browser-use: one decision-model request per action cycle drives a real browser, MIT
Yohei Nakajima's autonomous task agent - now a self-building agent framework in Python
Gratuit et ouvert : récupérez le code source, mettez une étoile au dépôt et commencez à créer.