# Openmesha

> Openmesha is an open-source MCP server for agent tooling. Open-source agent operations mesh with workflows, budgets, governance and an MCP server. It is available on API, CLI and self-hosted.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** API, CLI, self-hosted
- **Repository:** https://github.com/anamized/openmesha
- **MCP Registry name:** io.github.ANAMIZED/openmesha
- **Categories:** [agent tooling](https://ai-powered.com/categories/agent-tooling)
- **Listing updated:** 2026-11-18

## About

The README describes OpenMesha as an agentic operating system with two parts: a web control plane built around a routing and memory kernel, and a Python control plane that treats agents as processes with budgets. Agents can be grouped into multi-agent workflows, and cost is tracked as a first-class resource alongside governance checks that fail closed.

According to the README, the project ships several surfaces: a REST API with OpenAPI docs, a CLI, an MCP server (openmesha-mcp), a Python SDK, and skill definitions for coding agents. It is started with Docker Compose and checked with a bundled verification script.

The project targets engineers who want to run and inspect an agent control plane themselves. The README also lists non-custodial USDC addresses and x402-related payment features, and it links to a separate production path document for cloud or self-hosted deployment.

## Features

- **Multi-agent workflows:** Runs groups of agents through workflows exposed via the REST API, MCP, CLI and skills.
- **Budget and cost tracking:** Treats cost as a managed resource, with budgets attached to agents as processes.
- **Governance guardrails:** Uses AgentFence and HOTL checks that the README says fail closed by default.
- **MCP server:** Exposes the control plane to MCP-compatible AI clients via the openmesha-mcp command.
- **Docker-based deployment:** Brings up the API and services with a single docker compose command and a verification script.

## Good for

- Coordinating multiple agents on a shared workflow with cost limits
- Testing agent governance rules before running them in production
- Driving an agent control plane from scripts with the Python SDK or CLI
- Exposing agent operations to an MCP-compatible assistant

## FAQ

### How do I start OpenMesha?

The README's quick start runs docker compose up --build and then bash scripts/verify.sh, with the API served on port 8080.

### What license is it under?

The README states Apache-2.0.

### Where is the API documentation?

The README lists OpenAPI docs at http://localhost:8080/docs when the stack is running.

### Is Openmesha free?

Yes. Openmesha is open source, so you can use and self-host it for free.

### What are the best alternatives to Openmesha?

The closest alternatives to Openmesha are Server Os, Yodmcp, mcp-gateway-registry, Synapbus and Agentome.

## Where this comes from

- Description: written from the project's README and its MCP Registry entry.

## Alternatives

- [Server Os](https://ai-powered.com/mcp/server-os): Server Os is an open-source MCP server for agent tooling.
- [Yodmcp](https://ai-powered.com/mcp/yodmcp): Yodmcp is an open-source MCP server for agent tooling.
- [mcp-gateway-registry](https://ai-powered.com/mcp/mcp-gateway-registry): mcp-gateway-registry is an open-source MCP server for agent tooling.
- [Synapbus](https://ai-powered.com/mcp/synapbus): Synapbus is an open-source MCP server for agent tooling.
- [Agentome](https://ai-powered.com/mcp/agentome): Agentome is an open-source MCP server for agent tooling.
- [Oauth Multi Llm Orchestrator](https://ai-powered.com/mcp/oauth-multi-llm-orchestrator): Oauth Multi Llm Orchestrator is an MCP server for ai models and llm tooling.
- [Superagenticmcp](https://ai-powered.com/mcp/superagenticmcp): Superagenticmcp is an open-source MCP server for agent tooling.
- [OpenCrawl](https://ai-powered.com/mcp/opencrawl): OpenCrawl is a freemium MCP server for agent tooling.

---
Source: https://ai-powered.com/mcp/openmesha
