AgentServe

A simple and efficient tool to deploy OpenAI Agents as FastAPI web services.

Overview

AgentServe is a lightweight wrapper that allows you to easily serve OpenAI Agents as web services. It provides a simple API for interacting with agents, supporting both synchronous and streaming responses.

Installation

pip install openai-agentserve

Or with Poetry:

poetry add openai-agentserve

Requirements

• Python 3.8+
• FastAPI
• Uvicorn
• OpenAI Agents SDK

Quick Start

from agents import Agent
from agentserve import serve

# Initialize your agent
my_agent = Agent(name="Assistant", instructions="You are a helpful assistant")

# Serve the agent on port 8000
serve(agent=my_agent, port=8000)

Usage Examples

Invoking the API

The API supports both streaming and non-streaming responses.

Streaming Response

import requests

url = "http://localhost:8000/invoke"

response = requests.post(url, json={
    "input": "please tell me a about the history of the world",
    "stream": True
    }, stream=True)
for i, line in enumerate(response.iter_lines()):
    if line:
        decoded_line = line.decode('utf-8').strip()
        print(decoded_line)

Non-Streaming Response

import requests

url = "http://localhost:8000/invoke"

response = requests.post(url, json={
    "input": "please tell me about the history of the world",
    "stream": False
})
result = response.json()
print(result)

API Documentation

AgentServe provides interactive API documentation through two interfaces:

ReDoc Documentation

Access the ReDoc documentation at /redoc when your server is running:

http://localhost:8000/redoc

ReDoc provides a clean, organized interface with:

• Detailed endpoint descriptions
• Request/response examples
• Interactive API testing
• Schema documentation
• Endpoint grouping by tags

Swagger UI Documentation

Access the Swagger UI documentation at /docs:

http://localhost:8000/docs

Swagger UI provides:

• Interactive API testing
• Request/response visualization
• Schema exploration
• Try-it-out functionality

OpenAPI Schema

The OpenAPI schema is available at /openapi.json:

http://localhost:8000/openapi.json

This can be used with any OpenAPI-compatible tools or documentation generators.

Development

For information about CI/CD pipelines and GitHub Actions workflows, see .github/workflows/README.md.

Development Workflow

1. Fork the repository
2. Create a feature branch:

   git checkout -b feature/your-feature-name
   

3. Make your changes
4. Run local checks:

   poetry install
   poetry run pytest
   poetry run black .
   poetry run isort .
   poetry run flake8
   

5. Commit your changes:

   git add .
   git commit -m "feat: your feature description"
   

6. Push to your fork:

   git push origin feature/your-feature-name
   

7. Create a Pull Request

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.