Antigravity CLI — Tutorials

Step-by-step tutorials for Antigravity CLI. Setting up MCP servers, installing plugins, automating PR reviews, and building agent skills.

June 24, 2026
antigravityagytutorialsmcppluginsautomation

Antigravity CLI — Tutorials

Step-by-step guides for common Antigravity CLI workflows. For skills-based workflows, see the Agent Skills guide.

Setting Up MCP Servers

Connect Antigravity CLI to databases, APIs, and external services:

# Install a Postgres MCP server
agy mcp add postgres npx @modelcontextprotocol/server-postgres

# Install a GitHub MCP server
agy mcp add github npx @modelcontextprotocol/server-github

# Verify
agy mcp list

Now your agent can query databases and interact with GitHub:

> Query the users table and show me the top 10 by signup date
> List my open pull requests on this repo

Installing a Plugin

Extend Antigravity CLI with community plugins:

# Install a security audit skill
agy plugin install https://github.com/user/security-audit-skill

# List installed plugins
agy plugin list

# List available skills
> /skills

Automating PR Reviews

Set up automatic code review with GitHub Actions:

name: AI Review
on: [pull_request]
jobs:
  review:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - run: curl -fsSL https://antigravity.google/cli/install.sh | bash
      - run: agy -p "Review this PR for bugs, security, and style" \
          --output-format json \
          --model "Claude Opus 4.6 (Thinking)"
        env:
          ANTIGRAVITY_API_KEY: ${{ secrets.ANTIGRAVITY_API_KEY }}

Related Articles

Windsurf — AI Coding Tool Guide

Windsurf IDE with Cascade AI flow editor. Prompt patterns, rules configuration, multi-file editing, and comparison with Cursor and Copilot.

GitHub Copilot — AI Coding Assistant

Complete guide to GitHub Copilot, the AI coding assistant. Covers prompt engineering, configuration, copilot-instructions.md, agent mode, and GitHub workflow integration.

Prompt Engineering in Cursor

Master prompt engineering in Cursor IDE. Agent mode vs inline editing, composer prompts, codebase-aware prompting, and .cursorrules integration strategies.