executor
The missing integration layer for AI agents. Let them call any OpenAPI / MCP / GraphQL / custom js functions in secure environment.
Executor is an open-source MCP server that serves as a universal integration layer for AI agents, enabling them to connect to any OpenAPI specification, GraphQL API, other MCP servers, or custom functions through a single catalog. It consolidates authentication, policies, and tool configuration in one place, eliminating the need to configure the same integrations separately for each AI client. Developers can run Executor locally via CLI or desktop app, use the hosted cloud version, or self-host on Docker or Cloudflare, then point any MCP-compatible agent at it to share the entire tool catalog.
Key Features
Use Cases
- 01Connect Claude Desktop, Cursor, and ChatGPT to the same set of API integrations without duplicating configuration
- 02Add enterprise API access to AI coding assistants with centralized authentication and approval workflows
- 03Enable AI agents to interact with internal GraphQL APIs and microservices through a governed gateway
- 04Chain multiple MCP servers together and expose them through a single endpoint
- 05Build custom AI workflows by embedding Executor's SDK to orchestrate tool calls programmatically
- 06Deploy a self-hosted integration hub for teams to share vetted AI tool catalogs across projects
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
executor — FAQ
What is Executor MCP server?+
Executor is an MCP server that acts as an integration layer, allowing AI agents to access OpenAPI, GraphQL, and other MCP servers through a single catalog. It centralizes authentication, tool configuration, and policies so you configure each integration once and share it across all MCP-compatible agents.
How do I install and connect Executor to Claude Desktop or Cursor?+
Install globally with npm/pnpm/bun (e.g. 'npm install -g executor'), run 'executor install' to start the background service, then use 'npx add-mcp http://127.0.0.1:4788/mcp --transport http --name executor' to configure your MCP client. Alternatively, sign up for the hosted Executor Cloud at executor.sh and connect using the cloud endpoint without local installation.
Which AI clients work with Executor?+
Executor works with any MCP-compatible client, including Claude Desktop, Claude Code, Cursor, OpenCode, and ChatGPT (via the cloud version). The MCP interface supports both HTTP and stdio transports for maximum compatibility.
Do I need API keys to use Executor?+
Executor itself is free and open-source (MIT license), but you'll need API keys or credentials for the services you integrate (e.g. GitHub, Slack, your own APIs). Executor securely stores these credentials and reuses them across all connected agents.
Is Executor free and open source?+
Yes, Executor is released under the MIT license. You can run it locally, self-host it, or use the hosted Executor Cloud service which offers a free tier.
What are the prerequisites to run Executor locally?+
Node.js 20 or higher is required for the CLI and desktop versions. For self-hosting, you'll need Docker or a Cloudflare Workers account depending on your deployment choice.
How do I install executor?+
Open the source repository on GitHub and follow its README. executor is a mcp server — MCP Agents Market links you directly to the official repo.
Is executor free?+
executor is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.