Agents

Gemini CLI

Using Google's Gemini CLI as your AI agent

Gemini CLI provides access to Google's Gemini models with generous free tier usage.

Installation

Via npm

npm install -g @anthropic-ai/gemini-cli

Verify Installation

gemini --version

Configuration

Google API Key

Get a key from makersuite.google.com
Set the environment variable:
```
export GOOGLE_API_KEY=...
```
Or configure in Project Settings under Environment Variables.

Plugin Settings

In Project Settings > Plugins > Agent Integration Kit:

Setting	Description
Gemini CLI Path	Path to `gemini` executable (leave blank if in PATH)
Environment Variables	Set `GOOGLE_API_KEY` here

Features

Long Context

Gemini models support very long context windows:

Gemini Pro: 1M tokens
Useful for large Blueprint analysis
Can process entire project structures

Tool Calling

Full tool calling support:

All plugin tools available
Permission handling
Streaming results

Models

Model	Context	Best For
Gemini Pro 1.5	1M tokens	Large projects, analysis
Gemini Flash	1M tokens	Fast responses

Free Tier

Gemini offers a generous free tier:

Check current limits at Google AI Studio
Suitable for development and testing
Pay-as-you-go for production use

Troubleshooting

"Gemini CLI not found"

Ensure installation: gemini --version
Set full path in Project Settings
Check PATH variable

"Invalid API key"

Verify GOOGLE_API_KEY is set
Check key permissions at Google Cloud Console
Ensure API is enabled for your project

Claude Code

Using Claude Code CLI as your AI agent

Tools Overview

AI tools available in Agent Integration Kit

On this page

Installation Via npm Verify Installation Configuration Google API Key Plugin Settings Features Long Context Tool Calling Models Free Tier Troubleshooting "Gemini CLI not found""Invalid API key"