openai2claude 0.1.0

Proxy Claude API requests to OpenAI-compatible backends

openai2claude

Proxy server that converts Claude API requests to OpenAI-compatible backends. This allows you to use the Anthropic Claude SDK with any OpenAI-compatible API provider.

Installation

pip install openai2claude

Quick Start

1. Start the proxy server:

export OPENAI2CLAUDE_BACKEND=https://api.openai.com/v1
export OPENAI2CLAUDE_API_KEY=sk-your-api-key
openai2claude

Or with command-line arguments:

openai2claude --backend https://api.openai.com/v1 --api-key sk-your-api-key

2. Configure your Claude SDK code to use the proxy:

import anthropic

client = anthropic.Anthropic(
    base_url="http://localhost:8000"
)

response = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello, world!"}]
)

Or via environment variable:

export ANTHROPIC_API_BASE="http://localhost:8000"

Configuration

Environment Variables

Variable	Default	Description
OPENAI2CLAUDE_BACKEND	http://localhost:8080/v1	OpenAI-compatible backend URL
OPENAI2CLAUDE_API_KEY	sk-xxx...	API key for the backend
OPENAI2CLAUDE_HOST	0.0.0.0	Proxy server host
OPENAI2CLAUDE_PORT	8000	Proxy server port
OPENAI2CLAUDE_LOG_LEVEL	INFO	Logging level
OPENAI2CLAUDE_TIMEOUT	300	Request timeout in seconds
OPENAI2CLAUDE_VERIFY_SSL	true	Verify SSL certificates

Command-Line Options

openai2claude [OPTIONS]

Options:
  -b, --backend TEXT         OpenAI-compatible backend URL
  -k, --api-key TEXT         API key for the backend
  -h, --host TEXT            Host to bind the proxy server
  -p, --port INTEGER         Port to bind the proxy server
  -l, --log-level TEXT       Logging level (debug, info, warning, error)
  -t, --timeout FLOAT        Request timeout in seconds
  --no-verify-ssl            Disable SSL certificate verification
  --version                  Show version

Architecture

┌─────────────┐    ┌─────────────────┐    ┌────────────────────┐
│  Your Code  │───>│  openai2claude  │───>│  OpenAI Backend    │
│ Claude SDK  │    │    (localhost   │    │  (api.openai.com,  │
│             │<───│     :8000)      │<───│  ollama, etc.)     │
└─────────────┘    └─────────────────┘    └────────────────────┘

The proxy:

1. Accepts requests in Claude API format (/v1/messages)
2. Converts them to OpenAI API format (/v1/chat/completions)
3. Forwards to the configured backend
4. Converts responses back to Claude format

Features

• Format Conversion: Automatically converts between Claude and OpenAI API formats
• Tool Calling Fix: Fixes malformed tool calls (e.g., cmd/input -> command) for Bash tools
• Streaming Support: Full SSE streaming support for real-time responses
• System Prompt Injection: Adds tool usage instructions to prevent common errors
• Environment Variables: Full configuration via environment variables
• CLI Interface: Easy-to-use command-line interface with --help

Use Cases

1. Use Claude SDK with local models: Point to Ollama, LM Studio, or other OpenAI-compatible local servers
2. Multi-provider support: Switch between OpenAI, Azure, or other OpenAI-compatible providers
3. Development and testing: Test Claude code against mock or local backends

License

MIT