fastapi-praisonai

FastAPI integration for PraisonAI multi-agent framework.

Installation

pip install fastapi-praisonai

Quick Start

from fastapi import FastAPI
from fastapi_praisonai import create_router

app = FastAPI()
app.include_router(create_router())

# Now you have:
# POST /praisonai/query - Send queries to PraisonAI
# GET /praisonai/agents - List available agents

API Endpoints

POST /praisonai/query

Send a query to PraisonAI agents:

{
  "query": "Research AI trends",
  "agent": "researcher"  // optional
}

GET /praisonai/agents

List available PraisonAI agents.

Using the Client Directly

from fastapi_praisonai import PraisonAIClient

client = PraisonAIClient(api_url="http://localhost:8080")

# In an async context
result = await client.run_workflow("Research AI trends")
result = await client.run_agent("Write an article", "writer")
agents = await client.list_agents()

Configuration

from fastapi_praisonai import create_router

router = create_router(
    api_url="http://localhost:8080",
    prefix="/ai",  # Custom prefix
    tags=["AI Agents"],  # OpenAPI tags
)

Prerequisites

Start PraisonAI server:

pip install praisonai
praisonai serve agents.yaml --port 8080

Links

• PraisonAI Documentation
• FastAPI Documentation

License

MIT