Antigravity MCP Bridge

Antigravity MCP Bridge

A TypeScript-based MCP server template for the Antigravity agentic environment, providing system status and SQLite database tools via stdio transport.

Category
Visit Server

README

Antigravity MCP Bridge

A TypeScript-based Model Context Protocol (MCP) server template wired into the Antigravity agentic environment. It exposes tools over a stdio transport and ships with a ready-to-import SQLite skill for local database access.


Prerequisites

Requirement Version
Node.js ≥ 18
npm ≥ 9
Antigravity latest

Project Structure

antigravity-mcp-bridge/
├── src/
│   └── server.ts          # MCP server — stdio transport + tools
├── skills/
│   └── sqlite_skill.json  # Antigravity Global Skill definition
├── dist/                  # Compiled JS output (after build)
├── auto_bridge.sh         # Detects MCP configs & writes env settings
├── tsconfig.json
├── package.json
└── README.md

1 · Starting the MCP Server

Build & run (recommended)

# Install dependencies (first time only)
npm install

# Compile TypeScript → dist/
npm run build

# Start the server
node dist/server.js

The server writes a startup message to stderr and listens for MCP messages on stdin/stdout.

Development mode (ts-node)

npx ts-node src/server.ts

Auto-bridge setup

Run auto_bridge.sh once after cloning (or whenever your MCP config changes). It scans well-known config locations, detects your Node environment, and writes .env.mcp plus a descriptor into ~/.antigravity/mcp/.

chmod +x auto_bridge.sh   # already done if you followed setup
./auto_bridge.sh

2 · Importing the SQLite Skill into Antigravity Manager

  1. Open the Antigravity desktop app and navigate to the Manager View (sidebar → 🧩 Manager).
  2. Click Import Skill (top-right of the Skills panel).
  3. In the file picker, navigate to:
    /path/to/antigravity-mcp-bridge/skills/sqlite_skill.json
    
    and select it.
  4. Antigravity will validate the skill schema and display a "SQLite Database Connector" card under the Data Sources category.
  5. Click Configure on the card and fill in the required field:
    • SQLite Database Path — absolute path to your .db / .sqlite file, e.g. /Users/sameer/data/myapp.db
  6. Click Save & Activate. The skill is now globally available to all agents.

Tip: You can activate / deactivate the skill at any time from the Manager View without removing the import.


3 · Testing the Connection with the Built-in Browser

Antigravity ships a built-in MCP Inspector browser that lets you call tools interactively.

Step-by-step

  1. Start the MCP server (see §1 above) so it is listening.
  2. In Antigravity, open Settings → MCP Servers and click + Add Server.
  3. Set:
    • Transport: stdio
    • Command: node /path/to/antigravity-mcp-bridge/dist/server.js
  4. Click Connect. The status indicator should turn green.
  5. Navigate to Tools tab inside the MCP Inspector.
  6. Select the system_status tool, leave the input blank (no parameters), and click Run.
  7. The response pane should display a JSON payload similar to:
    {
      "currentTime": "2026-04-23T02:19:15.000Z",
      "platform": "darwin",
      "release": "24.4.0",
      "architecture": "arm64",
      "hostname": "MacBook-Pro",
      "cpus": 10,
      "totalMemoryMB": 16384,
      "freeMemoryMB": 4096,
      "uptime": "3h 42m",
      "nodeVersion": "v22.0.0"
    }
    
  8. To test the SQLite skill, ensure the skill is imported and configured (§2), then ask an agent: "List all tables in my database." The agent will invoke sqlite_list_tables via the MCP bridge automatically.

Available Tools

Tool Description
system_status Returns current UTC time and OS details
sqlite_query Execute a read-only SQL SELECT
sqlite_execute Execute a write SQL statement
sqlite_list_tables List all tables in the SQLite DB
sqlite_describe_table Describe columns of a table

Scripts

npm run build    # tsc — compile src/ → dist/
npm run dev      # ts-node src/server.ts (watch mode)
npm run start    # node dist/server.js

Adding More Tools

Open src/server.ts and call server.tool(name, description, inputSchema, handler):

server.tool(
  "my_new_tool",
  "Does something useful.",
  { param: z.string().describe("An example parameter") },
  async ({ param }) => ({
    content: [{ type: "text", text: `You passed: ${param}` }],
  })
);

Rebuild with npm run build and reconnect in Antigravity.


License

MIT

Created by:

Sameer Abrar, Flexcrit Inc

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