Terminal Hook
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Terminal Hookcheck the output from my zsh terminal for any error messages"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Terminal Hook
A VSCode/Cursor extension that captures terminal output in real-time and exposes it via MCP (Model Context Protocol) for AI assistants.
Why another extension?
copy-pasting terminal output and manually referencing it in every chat is so meehhh. When AI assistants can proactively read your terminal — checking errors, logs, and command output without you having to break the execution to ask for more info, this extension solves that! Other MCP tools can also tap into terminal data for automation
Related MCP server: VibeWatch
How It Works
┌─────────────────┐ stdio ┌─────────────────┐ TCP ┌─────────────────┐
│ Cursor/Claude │◄──────────────►│ mcp-server.mjs │◄────────────►│ VSCode Extension│
│ (MCP Client) │ │ (MCP Server) │ port 9876 │ (Terminal Data) │
└─────────────────┘ └─────────────────┘ └─────────────────┘VSCode Extension captures terminal output via
onDidWriteTerminalDataAPIMCP Server (
mcp-server.mjs) communicates with Cursor via stdioTCP Bridge connects the MCP server to the extension on port 9876
Features
Real-time Capture - Captures all terminal output as it happens
Clean Output - Automatically strips ANSI codes and shell noise
MCP Integration - Exposes terminal data to AI assistants like Claude
Installation
See INSTALL.md for setup instructions.
MCP Tools
list_terminals
List all active terminals with metadata.
{
"success": true,
"count": 2,
"terminals": [
{
"id": "terminal-0-zsh",
"name": "zsh",
"processId": 1234,
"bufferLines": 150,
"lastActivity": "2024-12-18T10:30:00.000Z"
}
]
}get_terminal_output
Get output from a specific terminal.
Parameters:
terminal_name(required): Terminal name or IDlines(optional): Number of lines to return (default: 100)
{
"success": true,
"terminal": "zsh",
"output": "$ npm start\nServer running on port 3000\n...",
"lines_returned": 50
}Development
bun install
bun run compile
bun testLicense
MIT
This server cannot be deployed
Maintenance
Related MCP Connectors
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
Real-time chat for AI agents. Claude Code, Cursor, Cline and Codex join channels over MCP.
Persistent memory for Claude Code and Cursor. Stop re-explaining your project every session.
Let ChatGPT, Claude & Cursor use your Mac: email, calendar, iMessage, Teams, files. Local, free.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables AI assistants to fully interact with console applications, monitor output, detect errors, and automate terminal workflows across multiple sessions. Similar to how Playwright works for web browsers but for command-line interfaces.138 npm47MIT
- AlicenseNot gradedqualityDmaintenanceMonitors development commands and exposes terminal output to Claude in real-time, allowing AI assistants to see errors, logs, and stack traces without copy-pasting.4 npm3MIT
- AlicenseNot gradedqualityDmaintenanceExposes VS Code diagnostic problems like errors and warnings in real-time to AI agents via the Model Context Protocol. It enables AI tools to see compilation and linting issues directly from the problems panel to provide context-aware debugging and automated fixes.22MIT
- FlicenseNot gradedqualityAmaintenanceExposes VS Code terminals as MCP tools, allowing AI agents to create, execute commands, read output, list, and kill terminals across multiple workspaces.-