Gemini CLI - NagaAI Documentation

Gemini CLI can work with NagaAI through a community fork that supports OpenRouter-style configuration. This is not the official Gemini CLI path. Use it only if you specifically want Gemini CLI and are comfortable relying on a community-maintained fork.

This integration depends on a community fork or branch, not the official Gemini CLI release. Review the repository before using it in production.

Setup

Clone the OpenRouter-compatible fork/branch

Use the OpenRouter-compatible fork or branch instead of the default upstream release:

git clone https://github.com/heartyguy/gemini-cli
cd gemini-cli
git checkout feature/openrouter-support

Configure environment variables for NagaAI

Point the fork at NagaAI’s OpenAI-compatible endpoint:

export OPENROUTER_BASE_URL="https://api.naga.ac/v1"
export OPENROUTER_API_KEY="YOUR_NAGAAI_API_KEY"

Need an API key? Create one in the NagaAI Dashboard, then see Quickstart if you want a first working request.

Choose a tools-capable model

If you want repo search, file edits, or command execution, choose a model that supports function calling:

Browse models with tools support

Install dependencies and start the CLI

Requirements: Node.js 18+

npm install
npm start

Verify the setup

Ask the CLI to inspect the current project or explain a file. If it answers normally and can use tools when needed, the NagaAI connection is working.

Try it

“Summarize the project structure and key modules.”
“Find the code responsible for X and explain how it works.”
“Propose a refactor plan for module Y and list the files to change.”

Notes

Base URL: https://api.naga.ac/v1
Auth: standard NagaAI API key
This setup may drift as the community fork changes

​Setup

​Try it

​Notes

Setup

Try it

Notes