johnlindquist/mcp-shell-yeah
If you are the rightful owner of mcp-shell-yeah and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
mcp-shell-yeah is a Model Context Protocol server designed to manage long-running local development tasks like servers, build watchers, or test runners, providing a standardized MCP interface.
mcp-shell-yeah is a Model Context Protocol (MCP) server that addresses the challenge of managing long-running local development tasks such as servers, build watchers, or test runners. These tasks often need to persist across multiple conversations, and checking their status or logs can be cumbersome. mcp-shell-yeah provides a standardized MCP interface, enabling AI agents to reliably start, monitor, retrieve logs from, and stop these crucial shell processes, thereby streamlining AI-driven development workflows. It offers features like starting and managing long-running shells, maintaining shell state across conversations, checking real-time status, accessing shell output logs, and stopping specific or all managed shells. The server is built using Node.js and TypeScript, and it strictly follows the official MCP standard for tool responses and type usage.
Features
- Start & Manage Long-Running Shells: Launch and manage dev servers, test runners, build scripts, and more.
- Maintain Shell State Across Conversations: Shells persist and can be checked or controlled in future chats.
- Check Real-time Status: Instantly know if a process is running, crashed, or stopped.
- Access Shell Output Logs Reliably: Retrieve recent or full logs for any managed shell.
- Stop Specific Shells or All Managed Shells: Gracefully or forcefully terminate processes as needed.
