Back to Apps

Shell MCP
by zatevakhin
Secure shell command execution MCP server with advanced whitelisting and safety controls.
2 stars
Updated 4 months ago
Works in:claude
Exposes:Tools
What it does\nShell MCP provides a secure bridge for LLM agents to execute shell commands on the host machine. It implements a strict security model using Abstract Syntax Tree (AST) parsing to validate commands, preventing dangerous injections and ensuring only approved operations are performed.\n\n## Tools\n- shell: Executes a validated shell command. It supports complex security policies, including explicit whitelists (e.g., ls, grep) and blacklists (e.g., rm, sudo), and can require user confirmation for non-whitelisted actions.\n\n## Installation\nAdd the following to your claude_desktop_config.json:\njson\n{\n "mcpServers": {\n "shell-mcp": {\n "command": "shell-mcp"\n }\n }\n}\n\n\n## Supported hosts\nClaude Desktop
Quick install
cargo install --git https://github.com/zatevakhin/shell-mcp.gitInformation
- Pricing
- free
- Published
- 6/13/2026
- Updated
- 4 months ago
- stars
- 2
Categories
Choose your AI client and follow the steps below.
Claude Desktop
{
"mcpServers": {
"shell-mcp": {
"command": "shell-mcp"
}
}
}





