tumf/mcp-shell-server

Overview

A secure shell command execution server that implements the Model Context Protocol (MCP). It allows AI agents to run shell commands through a controlled interface, sending commands as MCP requests and receiving results.

Best for

Best for
Developers who need a secure, auditable bridge for AI agents to run shell commands on servers

Use cases

• Run shell scripts and commands from an AI assistant within a sandboxed environment
• Automate system administration tasks by delegating command execution to an agent
• Enable AI-driven file operations and process management on a remote server

How to use

Install

npx -y @smithery/cli install mcp-shell-server --client claude

Tools exposed

• ALLOW_PATTERNS
• MCP_SHELL_DEFAULT_TIMEOUT_SECONDS
• MCP_SHELL_MAX_TIMEOUT_SECONDS
• MCP_SHELL_OUTPUT_LIMIT_BYTES
• MCP_SHELL_CHILD_ENV_ALLOWLIST
• MCP_SHELL_SAFE_PATH
• execution_time

Tested with

Claude Desktop

Notes

A secure shell command execution server that implements the Model Context Protocol (MCP). It allows AI agents to run shell commands through a controlled interface, sending commands as MCP requests and receiving results.

175 stars on GitHub. Last updated 2026-04-02. Licensed MIT.

Use cases

• Run shell scripts and commands from an AI assistant within a sandboxed environment
• Automate system administration tasks by delegating command execution to an agent
• Enable AI-driven file operations and process management on a remote server

Pros

• Provides a structured, protocol-based way to execute shell commands from AI agents
• Written in Python, making it easy to audit and extend
• Sandboxed execution reduces risk of unintended system changes

Cons

• Limited to shell commands it knows about; custom commands require configuration
• Still requires careful permission management to avoid security gaps
• Relatively small community (175 stars) and early-stage project

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.