mcp-shell and shellguard
These two tools are competitors, both providing an MCP server to give LLMs shell access, with "sonirico/mcp-shell" offering more general secure and auditable command execution while "fawdyinc/shellguard" specifically focuses on read-only access over SSH.
About mcp-shell
sonirico/mcp-shell
Give hands to AI. MCP server to run shell commands securely, auditably, and on demand.
This project allows developers to connect an AI agent to a controlled shell environment, enabling the AI to execute specific commands on a server. It takes commands from an AI tool and returns the execution results, while providing security configurations to limit what the AI can do. This is designed for AI developers or system administrators who integrate AI agents into their workflows and need to manage their interaction with the underlying system securely.
About shellguard
fawdyinc/shellguard
MCP server that gives LLM agents read-only shell access over SSH
ShellGuard helps IT operations, site reliability engineers, and system administrators automate server diagnostics and troubleshooting. It lets your AI assistant securely connect to remote servers (development, staging, or production), run observation commands, inspect logs, and query databases. The AI gets read-only access to terminal output, eliminating the need for manual copy-pasting and speeding up issue resolution.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work