Gemini CLI GitHub Actions: Google Cloud's Native AI for Your DevOps Workflow

Integrate Google’s Gemini AI into your development workflow for autonomous coding, issue triage, and pull request reviews

Overview

run-gemini-cli is a GitHub Action that brings the power of Google’s Gemini AI directly into your GitHub repositories. It acts as both an autonomous agent for critical routine coding tasks and an on-demand collaborator you can quickly delegate work to.

After testing both Claude and Gemini GitHub Actions extensively, I’ve found Gemini CLI offers unique advantages, especially for teams already in the Google ecosystem. Here’s everything you need to know to get started.

Why Gemini CLI?

Google Cloud Native Advantages

1. Google Cloud Integration: Seamlessly works with GCP services
2. Free Tier: Generous free quotas from Google AI Studio
3. Dual Deployment Options:
   • Quick start with API keys
   • Enterprise-grade with Vertex AI integration
4. Full Stack GCP: Part of Google’s comprehensive AI ecosystem
5. Built-in Tools: Native GitHub CLI integration for advanced operations
6. Cost-Effective: Significantly cheaper for high-volume usage

Features

• Automation: Trigger workflows based on events (e.g., issue opening) or schedules (e.g., nightly)
• On-demand Collaboration: Trigger workflows in issue and pull request comments by mentioning @gemini-cli
• Extensible with Tools: Leverage Gemini’s tool-calling capabilities to interact with CLIs like GitHub CLI (gh)
• Customizable: Use a GEMINI.md file for project-specific instructions

Quick Start Guide

Step 1: Get Your Gemini API Key

1. Visit Google AI Studio
2. Click “Create API Key”
3. Copy your key (starts with AIza...)

Step 2: Add as GitHub Secret

1. Go to your repository’s Settings > Secrets and variables > Actions
2. Click New repository secret
3. Name: GEMINI_API_KEY
4. Value: Your API key from Step 1

Step 3: Setup Workflows

Option A: Automated Setup (Recommended)

# Install Gemini CLI globally
npm install -g @google/gemini-cli

# Start the CLI
gemini

# In the chat interface, type:
/setup-github

This automatically creates optimized workflow files for your repository.

Option B: Manual Setup

Create .github/workflows/gemini.yml:

name: Gemini CLI Assistant

on:
  issues:
    types: [opened, edited]
  issue_comment:
    types: [created, edited]
  pull_request:
    types: [opened, edited, synchronize]
  pull_request_review_comment:
    types: [created, edited]

jobs:
  gemini-assist:
    if: contains(github.event.comment.body || github.event.issue.body || github.event.pull_request.body, '@gemini-cli')
    runs-on: ubuntu-latest
    
    steps:
      - uses: actions/checkout@v4
      
      - name: Run Gemini CLI
        uses: google-github-actions/run-gemini-cli@v1
        with:
          api_key: $
          github_token: $
          model: gemini-1.5-flash  # or gemini-1.5-pro for complex tasks

Step 4: Test Your Setup

Create a test issue with:

@gemini-cli create a simple Python script that prints "Hello from Gemini!"

Within seconds, Gemini will:

1. Analyze your request
2. Create a new branch
3. Generate the code
4. Open a pull request

Workflow Examples

Pull Request Review

name: Automated PR Review

on:
  pull_request:
    types: [opened, synchronize]

jobs:
  review:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      
      - name: Gemini Review
        uses: google-github-actions/run-gemini-cli@v1
        with:
          api_key: $
          github_token: $
          command: |
            Review this pull request for:
            - Code quality and best practices
            - Security vulnerabilities
            - Performance implications
            - Test coverage

Issue Triage

name: Issue Triage

on:
  issues:
    types: [opened]

jobs:
  triage:
    runs-on: ubuntu-latest
    steps:
      - name: Triage with Gemini
        uses: google-github-actions/run-gemini-cli@v1
        with:
          api_key: $
          github_token: $
          command: |
            Triage this issue:
            - Add appropriate labels
            - Suggest priority (P0-P3)
            - Identify related issues
            - Propose initial solution approach

Scheduled Documentation Updates

name: Weekly Doc Updates

on:
  schedule:
    - cron: '0 9 * * 1'  # Every Monday at 9 AM

jobs:
  update-docs:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      
      - name: Update Documentation
        uses: google-github-actions/run-gemini-cli@v1
        with:
          api_key: $
          github_token: $
          command: |
            Update README.md with:
            - Any new dependencies
            - Updated API documentation
            - Recent configuration changes

Advanced Configuration

Custom Instructions with GEMINI.md

Create a GEMINI.md file in your repository root:

# Project Context for Gemini

## Tech Stack
- Frontend: React with TypeScript
- Backend: Node.js with Express
- Database: PostgreSQL
- Testing: Jest and Cypress

## Coding Standards
- Use functional components with hooks
- Follow ESLint configuration
- Write tests for all new features
- Use conventional commits

## Project-Specific Rules
- Never modify the database schema without migration files
- All API endpoints must have OpenAPI documentation
- Performance: Keep bundle size under 200KB

Model Selection

Choose the right model for your needs:

# For simple tasks (faster, cheaper)
model: gemini-1.5-flash

# For complex analysis (slower, more capable)
model: gemini-1.5-pro

# For vision tasks (analyzing screenshots/diagrams)
model: gemini-1.5-flash-vision

Tool Integration

Enable GitHub CLI for advanced operations:

- name: Gemini with GitHub CLI
  uses: google-github-actions/run-gemini-cli@v1
  with:
    api_key: $
    github_token: $
    enable_tools: true
    tools:
      - github-cli
      - npm
      - python

Real-World Use Cases

1. Automated Bug Fixes

@gemini-cli The login button is not working on mobile devices. 
The error appears in LoginComponent.tsx line 45. Please fix this issue.

Gemini will:

• Analyze the component
• Identify the mobile-specific issue
• Create a fix
• Add responsive tests

2. Code Generation

@gemini-cli Create a REST API endpoint for user profile management with:
- GET /api/profile/:id
- PUT /api/profile/:id
- Validation middleware
- Unit tests

3. Performance Optimization

@gemini-cli Analyze the Dashboard component for performance issues and optimize:
- Reduce re-renders
- Implement lazy loading
- Add memoization where appropriate

4. Security Audit

@gemini-cli /security-audit

Perform a security audit of this PR checking for:
- SQL injection vulnerabilities
- XSS attacks
- Authentication bypasses
- Sensitive data exposure

Comparison: Gemini vs Claude GitHub Actions

Feature	Gemini CLI	Claude Code
Setup Time	5 minutes	30+ minutes
Requirements	API key only	Google Cloud + Vertex AI
Free Tier	Generous	Limited
Cost	~$1.25/million tokens	~$15/million tokens
Response Time	2-5 seconds	5-10 seconds
Code Quality	Excellent	Excellent
Documentation	Good	Excellent
Tool Integration	Native	Via MCP

Best Practices

1. Clear Instructions

❌ @gemini-cli fix this
✅ @gemini-cli Fix the TypeError in utils/parser.js line 23 when parsing null values

2. Iterative Development

Break complex tasks into smaller requests:

@gemini-cli Step 1: Create the database schema for a blog system
@gemini-cli Step 2: Add the API endpoints for CRUD operations
@gemini-cli Step 3: Create the frontend components

3. Review Before Merge

Always review Gemini’s changes:

• Check for security issues
• Verify business logic
• Ensure code style compliance
• Run tests locally

4. Use Branch Protection

Configure branch protection rules:

• Require PR reviews
• Run CI/CD checks
• Prevent direct commits to main

Cost Optimization

Free Tier Limits

• Gemini 1.5 Flash: 15 RPM, 1 million TPM, 1,500 RPD
• Gemini 1.5 Pro: 2 RPM, 32,000 TPM, 50 RPD

Tips to Stay Within Limits

1. Use gemini-1.5-flash for routine tasks
2. Batch similar requests
3. Cache responses for repeated queries
4. Set up quotas and alerts

Troubleshooting

Common Issues and Solutions

Issue: Workflow doesn’t trigger

• Ensure @gemini-cli is mentioned exactly
• Check workflow file syntax
• Verify GitHub Actions is enabled

Issue: API quota exceeded

• Switch to gemini-1.5-flash
• Implement rate limiting
• Consider paid tier

Issue: Incorrect code generation

• Add project context in GEMINI.md
• Be more specific in requests
• Provide example code

Conclusion

Gemini CLI GitHub Actions offers the fastest path to AI-assisted development with minimal setup and generous free tier. It’s perfect for:

• Individual developers wanting AI assistance
• Open source projects needing automation
• Startups optimizing development speed
• Teams already using Google Cloud

The combination of ease of setup, cost-effectiveness, and powerful capabilities makes Gemini CLI an excellent choice for integrating AI into your GitHub workflow.

---

Get Started Today!

1. Get your API key
2. View the official repo
3. Follow me for more AI DevOps content

Questions? Found a cool use case? Share in the comments or reach out @AIwithTim!