executor (RhysSullivan/executor) is an open-source integration layer for agent tool-calling on GitHub. It provides one controlled runtime where AI agents can invoke OpenAPI endpoints, MCP servers, GraphQL operations, and custom JavaScript functions without rebuilding integration logic for every workflow.
License
MIT
Stars
1,948
Homepage
https://executor.sh/Features
- Unified entrypoint for OpenAPI, MCP, GraphQL, and JavaScript functions
- Tool-calling abstraction layer built for AI agents
- Executes external calls in a controlled secure runtime
- Supports custom JavaScript function extensions
- Can act as the integration middle layer for multi-agent workflows
- Open-source under MIT for self-hosting and customization
Use Cases
- Evaluate executor when the need is MCP and tool-calling integration and the repo summary matches: Integration layer for AI agents to call OpenAPI, MCP, GraphQL, and custom JavaScript fu...
- Compare executor's implementation approach before committing to an internal build.
- Use executor to connect tool-enabled agent workflows to the repository capability.
- Use executor to test agent coordination patterns with a concrete open-source codebase.
- Use executor to study developer-tooling implementation details before building internal workflows.
- Complete a MIT license review before packaging executor into a commercial or hosted workflow.
FAQ
Start from the repository summary (Integration layer for AI agents to call OpenAPI, MCP, GraphQL, and custom JavaScript functions in a secure runtime.), then verify maintenance status, integration boundaries, and whether its MCP and tool-calling integration, agent orchestration, developer engineering workflows focus matches the intended workflow. Repository: https://github.com/RhysSullivan/executor. Stars: about 1,191. License: MIT.
executor is best treated as a repository-level component or reference implementation for MCP and tool-calling integration, agent orchestration, developer engineering workflows. Good evaluation scenarios include: Evaluate executor when the need is MCP and tool-calling integration and the repo summary matches: Integration layer for AI agents to call OpenAPI, MCP, GraphQL, and custom JavaScript fu... Compare executor's implementation approach before committing to an internal build. Use executor to connect tool-enabled agent workflows to the repository capability.
