freshcrate
Skin:/
Language:English/įŽ€äŊ“中文
Home > MCP Servers > linux-mcp-server

linux-mcp-server

Tools to allow LLM clients to interact with Linux systems remotely

python
Why this rank:Strong adoptionRecent releaseHealthy release cadence

Description

Tools to allow LLM clients to interact with Linux systems remotely

README

CI Coverage PyPI Docs

Linux MCP Server

A Model Context Protocol (MCP) server for read-only Linux system administration, diagnostics, and troubleshooting on RHEL-based systems.

Features

  • Read-Only Operations: All tools are strictly read-only for safe diagnostics
  • Remote SSH Execution: Execute commands on remote systems via SSH with key-based authentication
  • Multi-Host Management: Connect to different remote hosts in the same session
  • Comprehensive Diagnostics: System info, services, processes, logs, network, and storage
  • Configurable Log Access: Control which log files can be accessed via environment variables
  • RHEL/systemd Focused: Optimized for Red Hat Enterprise Linux systems

Installation and Usage

For detailed instructions on setting up and using the Linux MCP Server, please refer to our official documentation:

  • Installation Guide: Detailed steps for pip, uv, and container-based deployments.
  • Usage Guide: Information on running the server, configuring LLM clients, and troubleshooting.
  • Cheatsheet: A reference for what prompts to use to invoke various tools.

Release History

VersionChangesUrgencyDate
1.7.0## What's Changed * Add tools for [Performance Co-Pilot](https://pcp.io/). (#568, #637, #638, @panyamkeerthana, @owtaylor) * Add LINUX_MCP_HOST_MODE=[local-only,remote-only,any] config setting to control where tools may run (#603, @owtaylor) * Except for LINUX_MCP_HOST_MODE=local-only, make the hostname parameter mandatory to avoid models accidentally running things on localhost. (#601, @owtaylor, reported by @lifetoldme) * docs: Add a guide in the documentation for configuring KeyclHigh10/2/2026
1.6.0## Breaking Changes In 1.6.0 we've dropped our dependency on LiteLLM in favor of directly implementing our own client for inference APIs. This greatly reduces our dependency stack and eases maintenance. Because of this, the way of configuring the gatekeeper model for [Guarded Command Execution](https://rhel-lightspeed.github.io/linux-mcp-server/model changed. Instead of setting: ```sh LINUX_MCP_GATEKEEPER__MODEL=openai/gpt-5.4 ``` Set: ```sh LINUX_MCP_GATEKEEPER__PROVIDER=openai LIHigh9/4/2026
1.5.0## What's Changed * fix(ci): try a better workaround for ETXTBUSY by @owtaylor in https://github.com/rhel-lightspeed/linux-mcp-server/pull/384 * ci: use upstream pull refs for Testing Farm requests by @alexxa in https://github.com/rhel-lightspeed/linux-mcp-server/pull/390 * docs: reorganize and simplify documentation by @owtaylor in https://github.com/rhel-lightspeed/linux-mcp-server/pull/388 * ci: add GitHub-GitLab PR mirroring integration by @owtaylor in https://github.com/rhel-lightspeed/linuHigh7/8/2026
1.4.1## What's Changed * Disable builds for macOS with Python 3.14t by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/378 * Increase test timeout by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/379 * test: migrate functional tests into monorepo and configure CI by @alexxa in https://github.com/rhel-lightspeed/linux-mcp-server/pull/328 * Add optional dependency for Google Cloud Platform by @JoseExposito in https://github.com/rhel-lightspeed/linux-mcp-serHigh4/10/2026
1.4.0## What's Changed * Conditionally use a Pydantic model as a response format by @subpop in https://github.com/rhel-lightspeed/linux-mcp-server/pull/252 * Further-simplify-response_format by @subpop in https://github.com/rhel-lightspeed/linux-mcp-server/pull/255 * [RSPEED-2335] Wrap executed commands in sudo+systemd-run where possible by @subpop in https://github.com/rhel-lightspeed/linux-mcp-server/pull/256 * Dynamically construct instructions by @subpop in https://github.com/rhel-lightspeed/linuHigh4/9/2026
1.3.2## What's Changed * Adjust MacOS installation notes by @owtaylor in https://github.com/rhel-lightspeed/linux-mcp-server/pull/277 * Return structured output from log tools by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/275 * Update doc string for execute_command by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/281 * Add jobs to create a release and build zip apps by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/282 Low2/19/2026
1.3.1## What's Changed * Use absolute path to mcp-publisher by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/268 **Full Changelog**: https://github.com/rhel-lightspeed/linux-mcp-server/compare/1.3.0...1.3.1Low2/12/2026
1.3.0## What's Changed * Add job for publishing to MCP registry by @r0x0d in https://github.com/rhel-lightspeed/linux-mcp-server/pull/211 * [RSPEED-2278] Extract common logic into _list_resources helper by @major in https://github.com/rhel-lightspeed/linux-mcp-server/pull/128 * Clarify how to run in a container by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/234 * Remove incorrect duplicate example from docs by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-serLow2/10/2026
1.2.1## What's Changed * Fix MCP configuration syntax errors by @major in https://github.com/rhel-lightspeed/linux-mcp-server/pull/218 * Add parameter examples for LLM guidance by @major in https://github.com/rhel-lightspeed/linux-mcp-server/pull/217 * Skip directories on different filesystems when listing directory size by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/221 * Improve field validation for path parameters by @samdoran in https://github.com/rhel-lightspeed/linLow1/20/2026
1.2.0## What's Changed * Add metadata in README for MCP registry to verify ownership by @r0x0d in https://github.com/rhel-lightspeed/linux-mcp-server/pull/205 * Add optional dependency for `gssapi` by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/206 * Add constraint for FastMCP by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/207 * Update docs-serve make target by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/209 * Make theLow1/19/2026
1.1.0## What's Changed * Update VS Code example configuration to use `mcp.json` by @x86girl in https://github.com/rhel-lightspeed/linux-mcp-server/pull/189 * Use `journalctl` for audit logs by @major in https://github.com/rhel-lightspeed/linux-mcp-server/pull/159 * Add instructions in the MCP server to provide better information to LLM clients by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/196 * Adjust code coverage requirements by @samdoran in https://github.com/rhel-lLow1/14/2026
1.0.1## What's Changed * Update project details by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/181 * Add functional testing to CI/CD by @kokesak in https://github.com/rhel-lightspeed/linux-mcp-server/pull/132 * Disable FastMCP startup banner by @samdoran in https://github.com/rhel-lightspeed/linux-mcp-server/pull/182 * Fix typos in README.md and system_info.py by @x86girl in https://github.com/rhel-lightspeed/linux-mcp-server/pull/183 * Bump asyncssh minimum to 2.22.0 fLow1/9/2026
1.0.0This is the initial release after several jam packed months of development. **Full Changelog**: https://github.com/rhel-lightspeed/linux-mcp-server/commits/1.0.0Low1/9/2026

Dependencies & License Audit

Loading dependencies...

Similar Packages

pipulateLocal First AI SEO Software on Nix, FastHTML & HTMXmain@2026-10-25
litellmPython SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropiv1.106.4
langfuse-mcpA Model Context Protocol (MCP) server for Langfuse, enabling AI agents to query Langfuse trace data for enhanced debugging and observabilityv0.14.0
rootly-mcp-serverRootly MCP servermain@2026-10-25
LIA-AssistantOpen-source multi-agent AI assistant powered by LangGraph, FastAPI & Next.js — 16+ agents, Human-in-the-Loop, MCP integration, voice TTS, RAG, 500+ metrics, 6 languages.v3.7.0

More in MCP Servers

litellmPython SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropi
ruflo🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade archit
@claude-flow/cliRuflo CLI - Enterprise AI agent orchestration with 60+ specialized agents, swarm coordination, MCP server, self-learning hooks, and vector memory for Claude Code
claude-plugins-officialOfficial, Anthropic-managed directory of high quality Claude Code Plugins.