A comprehensive plugin that provides OpenAI-compatible API endpoints for Cheshire Cat AI, enabling seamless integration with third-party applications that expect OpenAI API format.

• Complete Cat Pipeline Integration: All responses use the full Cheshire Cat pipeline including RAG, memory, hooks, and personality
• OpenAI API Compatibility: Fully compatible with OpenAI's chat completions API
• Streaming Support: Real-time streaming responses with Server-Sent Events (SSE)
• Rate Limiting: Configurable per-user rate limiting
• Authentication: Integrated with Cat's permission system
• Error Handling: Robust error handling with proper HTTP status codes
• No Fallbacks: Ensures all responses go through Cat's complete system

• Cheshire Cat AI v1.0.0 or later
• Python 3.8+

1. Download or clone this plugin into your Cat's plugins directory:

   cd /path/to/cheshire-cat/plugins
   git clone https://github.com/mc9625/cat_openai_api.git

2. Restart your Cheshire Cat instance

3. The plugin will automatically activate and be available at the custom endpoints

Access the plugin settings through the Cat admin panel:

All endpoints are available under the /custom prefix:

GET /custom/health

GET /custom/v1/models

POST /custom/v1/chat/completions

POST /custom/message

curl -X POST http://localhost:1865/custom/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "cheshire-cat",
    "messages": [
      {"role": "user", "content": "Hello, how are you?"}
    ],
    "user": "user123"
  }'

curl -X POST http://localhost:1865/custom/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "cheshire-cat",
    "messages": [
      {"role": "user", "content": "Tell me a story"}
    ],
    "stream": true,
    "user": "user123"
  }'

curl -X POST http://localhost:1865/custom/message \
  -H "Content-Type: application/json" \
  -d '{
    "text": "What is the weather like?",
    "user_id": "user123"
  }'

import openai

# Configure to use your Cat instance
openai.api_base = "http://localhost:1865/custom/v1"
openai.api_key = "dummy"  # Not used but required by SDK

response = openai.ChatCompletion.create(
    model="cheshire-cat",
    messages=[
        {"role": "user", "content": "Hello from Python!"}
    ],
    user="python_user"
)

print(response.choices[0].message.content)

const response = await fetch('http://localhost:1865/custom/v1/chat/completions', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    model: 'cheshire-cat',
    messages: [
      { role: 'user', content: 'Hello from JavaScript!' }
    ],
    user: 'js_user'
  })
});

const data = await response.json();
console.log(data.choices[0].message.content);

This plugin enables integration with any application that supports OpenAI API format:

• Chatbots and Voice Assistants
• Customer Service Platforms
• Content Management Systems
• Educational Platforms
• Business Intelligence Tools

The plugin respects Cat's built-in authentication and permission system:

• Read permissions required for model listing
• Write permissions required for chat completions
• Rate limiting protects against abuse
• Input validation prevents malicious requests

{
  "id": "chatcmpl-1234567890",
  "object": "chat.completion",
  "created": 1677649420,
  "model": "cheshire-cat",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! I'm the Cheshire Cat. How can I help you today?"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 15,
    "total_tokens": 27
  }
}

data: {"id":"chatcmpl-1234","object":"chat.completion.chunk","choices":[{"delta":{"content":"Hello"}}]}

data: {"id":"chatcmpl-1234","object":"chat.completion.chunk","choices":[{"delta":{"content":" there!"}}]}

data: [DONE]

1. No response received

   • Check that your Cat instance is running
   • Verify the endpoint URL is correct
   • Check Cat logs for errors

2. Rate limit exceeded

   • Reduce request frequency
   • Increase rate limit in plugin settings
   • Use different user IDs to distribute load

3. Timeout errors

   • Check if Cat's pipeline is processing correctly
   • Verify your prompt doesn't trigger infinite loops
   • Check Cat's memory and hook configurations

Enable debug mode in plugin settings to get detailed logs:

# Check Cat logs for detailed information
docker logs cheshire_cat_core

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

MIT License - see LICENSE file for details

This plugin is designed for Cheshire Cat AI, an open-source conversational AI framework.

• Issues: Report bugs and request features on GitHub
• Documentation: Visit the Cheshire Cat AI documentation
• Community: Join the Discord server for support and discussions

"We're all mad here. But now we're OpenAI-compatible mad!" 🎩✨

Made with ❤️ for the Cheshire Cat AI community