Loading connector details…
Loading connector details…
Choose a unique username to continue using AgentHotspot
by Sunwood-ai-labs • Automation & Orchestration
A secure MCP server that executes a configurable list of pre-approved shell commands and streams real-time output for agent workflows.
Run allowed repository or system commands (git, ls, npm, python, etc.) in a controlled, pre-approved manner to avoid command injection.
Integrate with desktop assistants like Claude Desktop to execute local tasks via an MCP server communicating over stdio.
Real-time streaming of command output and robust error handling for automation, debugging, or CI-like workflows while maintaining environment isolation.
This Model Context Protocol (MCP) server executes pre-approved commands securely, validating command prefixes and avoiding shell execution to minimize injection risks. It streams real-time command output over stdio and includes environment sanitization, error handling, and a configurable allowed-commands list. Designed for integration with agents or desktop assistants (e.g., Claude Desktop), it enables automated, constrained local command execution for debugging, CI tasks, and automation.
Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.