antigravity-terminal-mcp

antigravity-terminal-mcp

Windows-focused MCP server for terminal automation with persistent PowerShell sessions, live session logs, and VS Code integrated terminal bridge.

Category
Visit Server

README

Antigravity Terminal MCP

Windows-focused MCP server for terminal automation with persistent sessions and live session logs. Created by AATHI.

Features

  • Persistent PowerShell sessions (session_id, default: default)
  • Incremental stdout/stderr polling
  • Send follow-up input into running sessions
  • Live log file path returned for each session
  • Configurable monitor mode via ANTIGRAVITY_MONITOR_MODE
  • VS Code integrated terminal metadata returned for each session
  • MCP standard methods (tools/list, tools/call) with legacy aliases

MCP Tools

  1. run_command
  2. command_status
  3. send_input
  4. list_sessions
  5. session_visual_info
  6. kill_command

Publish-Ready One-Command Usage (From GitHub)

After pushing this project to your GitHub, users can seamlessly install and attach it to Claude using uvx (or the more robust python -m uv variant for Windows):

# Standard uvx (if in PATH)
claude mcp add antigravity-terminal-mcp uvx --from git+https://github.com/aathishwar-13/antigravity-terminal-mcp.git antigravity-terminal-mcp

# More robust Windows version (if uvx is not in PATH)
claude mcp add antigravity-terminal-mcp python -m uv tool run --from git+https://github.com/aathishwar-13/antigravity-terminal-mcp.git antigravity-terminal-mcp

(Note: If the code is nested in a subdirectory, simply append #subdirectory=folder_name to the git URL).

Local Run

uvx --from . antigravity-terminal-mcp

or

python server.py

Monitor Modes

By default, the server does not open an external monitor window (ANTIGRAVITY_MONITOR_MODE=none). This keeps monitoring in VS Code when you use the integrated terminal bridge.

If you explicitly want a separate PowerShell window for each session monitor, enable external mode:

$env:ANTIGRAVITY_MONITOR_MODE = "external"
python server.py

To force no external monitor window:

$env:ANTIGRAVITY_MONITOR_MODE = "none"
python server.py

When run_command or session_visual_info is called, the response now includes:

  • log_path
  • watch_command
  • monitor_mode
  • vscode_terminal
  • monitor_open (quick-open metadata for clickable/open actions)

monitor_open includes:

  • command_uri (VS Code command URI for antigravityTerminal.openMonitor)
  • markdown_link (prebuilt markdown clickable link for chat UIs)
  • integrated_terminal_command (copy/paste command to open live monitor in integrated terminal)
  • open_terminal_uri (opens a VS Code integrated terminal)
  • open_terminal_with_cwd_uri (opens integrated terminal in session cwd)
  • run_in_active_terminal_uri (runs monitor command in active integrated terminal)

The vscode_terminal object is designed for a VS Code extension to open the same session monitor inside an integrated terminal.

VS Code Integrated Terminal Bridge

A lightweight extension scaffold lives in vscode-integration/.

What it does

  • Opens an integrated PowerShell terminal in the current VS Code window
  • Runs the returned watch_command / vscode_terminal.command
  • Lets you paste either the full session_visual_info JSON or just the log_path
  • Shows a status-bar button: MCP Monitor (bottom-right) to pick and open any existing session monitor
  • Auto-detects session_logs/*.log create/change activity and shows an Open Monitor popup action

How to try it locally

  1. Open antigravity_mcp/vscode-integration in VS Code.
  2. Press F5 to launch the extension host.
  3. Run Antigravity: Open Monitor From Session Info JSON from the Command Palette.
  4. Paste the JSON returned by session_visual_info or run_command.

Where to click

  • Bottom-right VS Code status bar: click MCP Monitor and choose a session
  • On new session creation: click Open Monitor in the popup notification

This uses the VS Code extension API, which is the correct way to open an integrated terminal in the current editor window. A plain Python process cannot directly create a terminal tab inside VS Code by itself.

Quick Integrated Monitor Command (No Copy/Paste)

Use the helper script from a VS Code integrated terminal:

Set-Location d:\Projects\Client_projects\Velyx_Chatbot\antigravity_mcp
.\open-monitor.ps1 -SessionId live-agent

Optional short alias for the current terminal session:

Set-Alias agmwatch .\open-monitor.ps1
agmwatch -SessionId live-agent

This keeps monitoring inside VS Code and does not open an external terminal window.

Example MCP Config

{
  "mcpServers": {
    "antigravity-terminal-mcp": {
      "command": "python",
      "args": [
        "-m",
        "uv",
        "tool",
        "run",
        "--from",
        "git+https://github.com/aathishwar-13/antigravity-terminal-mcp.git",
        "antigravity-terminal-mcp"
      ]
    }
  }
}

Or


{
  "mcpServers": {
    "antigravity-terminal-mcp": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/AATHI/antigravity-terminal-mcp.git",
        "antigravity-terminal-mcp"
      ]
    }
  }
}

👨‍💻 About the Author

AATHI is a developer focused on building high-performance, developer-centric automation tools for the next generation of AI agents. With a specialization in bridging the gap between headless LLM execution and rich user-facing terminal interactions, AATHI created this tool to provide a portable, professional terminal experience beyond the standard VS Code integration limits.

Feel free to reach out for feedback or technical support via this repository!

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured