omcp — an open protocol for connecting AI models to tools and data. A community-driven hard fork of the Model Context Protocol. Build without permission.
Every revolution starts with a single act of defiance. &8212; Mahatma Gandhi
omcp is a community-driven hard fork of the Model Context Protocol specification. It exists to keep the protocol an open, permissionless standard: developed asynchronously, reviewed transparently on GitHub, and decided on technical merit rather than on closed-door approval.
- omcp specification - omcp protocol schema - omcp documentation
The schema is defined in TypeScript first, but made available as JSON Schema as well, for wider compatibility.
The specification is published as HTML and PDF — every version, plus every proposal — at enclawed.github.io/omcp. It is generated from this repository on every change to main; nothing there is written by hand. To build it yourself, run npm run build:spec. See Published Specification.
omcp is drop-in compatible with the broader Model Context Protocol ecosystem. The wire protocol, schema field names, method names, and protocol version identifiers are unchanged, so existing clients, servers, and SDKs interoperate without modification. The fork is in how the specification is developed and governed, not in what goes over the wire.
There is no sponsor to find, no working group to join, and no approval to request. A pull request is judged on four things: it follows good coding practices, it solves a real problem, it clearly states the problem it solves, and it ships working unit tests — along with the data needed to run them — that anybody can verify independently.
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/enclawed--omcp.html)
ChatGPT thinks. Codex works. Use ChatGPT as the planning brain while keeping the Codex harness.
0xwilliamortiz/openclaude-improvedruns anywhere. uses anything
XiaoPuOuO/openchatx-mcpTurn ChatGPT into a local agent runtime. Operate your computer, use local tools, discover MCP servers, and delegate work to your own models through one MCP connection.
NiazMorshed2007/jev-reviewLocal-first MCP plugin for continuous software-quality review by AI coding agents, powered by Jev.
0xsline/OpenChatCutLocal-first conversational AI video editor with a professional multi-track timeline, Agent Skills, MCP integration, and Remotion-powered rendering.
herry2059/project-os-for-codexOpen-source control plane for Codex projects: Git-backed context, visible agent progress, scoped MCP access, resumable work, and safe handoffs.
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.