flask-praisonai

Flask integration for PraisonAI multi-agent framework.

Installation

pip install flask-praisonai

Quick Start

from flask import Flask
from flask_praisonai import create_blueprint

app = Flask(__name__)
app.register_blueprint(create_blueprint())

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

if __name__ == "__main__":
    app.run()

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 flask_praisonai import PraisonAIClient

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

result = client.run_workflow("Research AI trends")
result = client.run_agent("Write an article", "writer")
agents = client.list_agents()

Configuration

from flask_praisonai import create_blueprint

bp = create_blueprint(
    api_url="http://localhost:8080",
    url_prefix="/ai",  # Custom prefix
)

Prerequisites

Start PraisonAI server:

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

Links

• PraisonAI Documentation
• Flask Documentation

License

MIT