The AI coding agent that works with 8+ LLM providers including OpenAI, Gemini, Anthropic, Grok, and access to 200+ models via OpenRouter.

Installation · Features · Providers · Documentation · Changelog

Hawk is a powerful, extensible CLI coding agent that brings the intelligence of modern LLMs directly to your terminal. Unlike locked-in alternatives, Hawk is provider-agnostic — use your own API keys, run local models, or switch between 8+ providers including access to 200+ models via OpenRouter.

Built for developers who want control without compromise.

# One command to start coding with AI
hawk

npm install -g hawk

brew install hawk

git clone https://github.com/GrayCodeAI/hawk.git
cd hawk
bun install
bun run build
npm link

# OpenAI
bun run profile:init -- --provider openai --api-key sk-xxx --model gpt-4o

# Anthropic Claude
bun run profile:anthropic -- --api-key sk-ant-xxx --model claude-3-5-sonnet-latest

# Local Ollama (Free, Private)
bun run profile:init -- --provider ollama --model llama3.2 --base-url http://localhost:11434

# Grok / xAI
bun run profile:grok -- --api-key xai-xxx --model grok-2

hawk

Ask Hawk to help with any task:

• "Create a React component for a login form"
• "Refactor this function to use async/await"
• "Debug why my API calls are failing"
• "Write tests for this module"

Hawk supports 8 official provider profiles plus any OpenAI-compatible endpoint.

Hawk's universal adapter works with any API that follows the OpenAI format:

• DeepSeek — Cost-effective coding
• Groq — Fast inference, free tier
• Together AI — Open-source models
• Azure OpenAI — Enterprise deployments
• LM Studio — Local GUI inference
• Custom endpoints — Your own API

Configure any compatible provider using the /config command inside Hawk, or edit ~/.hawk/provider.json directly:

{
  "openai_api_key": "your-key",
  "openai_base_url": "https://your-endpoint.com/v1",
  "active_model": "model-name"
}

See full provider list for more.

Hawk is built with a modular architecture:

• CLI Interface — Ink-based React terminal UI
• Tool System — 25+ extensible tools for code operations
• Provider Layer — Universal adapter for any LLM API
• Session Manager — Cost tracking, history, state management

See ARCHITECTURE.md for detailed diagrams and system design.

# Install dependencies
bun install

# Build project
bun run build

# Run in development mode
bun run dev

# Run tests
bun run smoke

# Validate environment
bun run doctor:runtime

# Quick local setup with Ollama
bun run dev:fast

This is the 1.x branch of Hawk with:

• ✅ Provider-agnostic architecture
• ✅ 25+ production-ready tools
• ✅ Smart provider routing with failover
• ✅ Real-time cost tracking
• ✅ MCP (Model Context Protocol) support
• ✅ LSP integration

Check releases for the latest updates.

Support Hawk's development:

Your sponsorship helps us maintain and improve Hawk.

We welcome contributions! Please read our Contributing Guide before submitting PRs.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

• 💬 Discord — Get help, share ideas
• 🐦 X/Twitter — Updates and announcements
• 🐛 GitHub Issues — Bug reports and feature requests

See SECURITY.md for our security policy and vulnerability disclosure process.

MIT License — see LICENSE for details.