# Server Os

> Server Os is an open-source MCP server for agent tooling. Agentic OS MCP server that runs agents as processes with cost limits and governance. It is available on CLI, API and self-hosted.

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

## About

The README describes Server OS as an operating system for agents, where each agent is handled as a process with its own cost budget, governance rules and observability. It ships as a REST API, a CLI, an SDK, an MCP server and an offline web control plane, so the same runtime can be driven from a browser, a terminal or an AI client.

The project defaults to a deterministic offline mock LLM, so it can be started with Docker Compose and checked without an API key. Setting a provider mode and an OpenAI key switches it to real models. The README says its checks cover the API, cost, governance, multi-agent workflows, SDK, CLI, MCP and skills.

It suits developers building multi-agent systems who want spending caps, least-privilege capabilities and audit visibility built into the runtime rather than added afterwards. The source is Apache-2.0 licensed and can be self-hosted.

## Features

- **Agents as processes:** Agents run as managed processes inside the operating system runtime.
- **Cost budgets:** Spending is treated as a first-class resource that can be capped per agent.
- **Governance and capabilities:** Capabilities are derived from declared intent to enforce least privilege.
- **Fail-closed behaviour:** The system is designed to deny actions when checks cannot be satisfied.
- **Observability by default:** Runtime activity is exposed for monitoring and inspection.
- **Multiple surfaces:** Accessible through REST API, CLI, Python SDK, MCP server and a web control plane.
- **Offline mock mode:** A deterministic mock LLM lets the system run and be verified without paid model access.

## Good for

- Running multi-agent workflows with enforced spending limits
- Auditing what agents are permitted to do before deploying them
- Prototyping agent orchestration offline with a deterministic mock model
- Exposing an agent runtime to MCP-compatible AI clients

## FAQ

### What license is Server OS released under?

The README states it is licensed under Apache-2.0.

### Does it need an API key to run?

No. The default mode is an offline mock LLM; setting SERVER_OS_LLM_MODE=openai with an OpenAI key enables real models.

### How is it started?

The README gives docker compose up --build, then bash scripts/verify.sh to run the checks.

### Is Server Os free?

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

### What are the best alternatives to Server Os?

The closest alternatives to Server Os are Openmesha, AIOS, Agentome, Superagenticmcp and Yodmcp.

## Where this comes from

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

## Alternatives

- [Openmesha](https://ai-powered.com/mcp/openmesha): Openmesha is an open-source MCP server for agent tooling.
- [AIOS](https://ai-powered.com/mcp/aios): AIOS is an MCP server for agent tooling.
- [Agentome](https://ai-powered.com/mcp/agentome): Agentome is an open-source MCP server for agent tooling.
- [Superagenticmcp](https://ai-powered.com/mcp/superagenticmcp): Superagenticmcp 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.
- [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.
- [Lrsi](https://ai-powered.com/mcp/lrsi): Lrsi 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.

---
Source: https://ai-powered.com/mcp/server-os
