Claude Code Proxy

Claude Code Proxy

A proxy service that translates between Anthropic's Claude API format and OpenAI-compatible API formats. Built with Hono framework on Bun runtime, deployable to Cloudflare Workers, Docker, or as a standalone CLI.

Features

- API Translation: Seamless format conversion between Claude and OpenAI APIs
- Message Normalization: Handles nested content arrays and tool call mapping
- Streaming Support: Both streaming and non-streaming response handling
- Multiple Deployment Options: Cloudflare Workers, Docker, or npm package
- CLI Interface: Standalone executable with version and help flags
- Model Routing: Dynamic model selection for reasoning vs completion tasks

Installation & Usage

NPM Package (Recommended)

Global Installation

On-demand Execution (npx/bunx)
You can also run the proxy without installation:

Using with Claude Code

Docker (Quick Start)

Environment File Configuration

Cloudflare Workers

After deployment, your proxy will be available at https://your-worker-name.your-subdomain.workers.dev

Using with Claude Code

Complete Setup Example

1. Deploy the proxy:

2. Set environment variables:

3. Test the deployment:

4. Use with Claude Code:

Development

Local Development

Build and Publish

Configuration

Environment Variables

- CLAUDE_CODE_PROXY_API_KEY - Bearer token for upstream API
- ANTHROPIC_PROXY_BASE_URL - Upstream API URL (default: https://models.github.ai/inference)
- REASONING_MODEL - Model for reasoning requests (default: openai/gpt-4.1)
- COMPLETION_MODEL - Model for completion requests (default: openai/gpt-4.1)
- REASONING_MAX_TOKENS - Max tokens for reasoning model (optional)
- COMPLETION_MAX_TOKENS - Max tokens for completion model (optional)
- REASONING_EFFORT - Reasoning effort level for reasoning model (optional, e.g.,…