Platform Python SDK

Clean public API exports for easy PraisonAI Platform integration

Platform Python SDK provides clean public API exports for seamless integration with PraisonAI Platform features.

Quick Start

1. Simple Import

   from praisonai_platform import PlatformClient, create_app
   
   # Create FastAPI app
   app = create_app()
   
   # Use HTTP client
   async with PlatformClient("http://localhost:8000") as client:
       await client.register("[e-mail]", "password")
   
2. With Authentication

   from praisonai_platform import PlatformClient
   
   # Client with token
   client = PlatformClient(
       base_url="http://localhost:8000",
       token="your-jwt-token"
   )
   
   workspaces = await client.list_workspaces()
   

---

How It Works

Component	Purpose	Usage
create_app()	FastAPI factory	Server deployment
PlatformClient	HTTP client	API integration
__version__	Package version	Version checking

---

API Exports

The package exports three main components:

create_app

from praisonai_platform import create_app

# Create FastAPI application instance
app = create_app()

# Run with uvicorn
if __name__ == "__main__":
    import uvicorn
    uvicorn.run(app, host="0.0.0.0", port=8000)

PlatformClient

from praisonai_platform import PlatformClient

# Initialize client
client = PlatformClient("http://localhost:8000")

# Authenticate
await client.register("[e-mail]", "pass")

# Create workspace
workspace = await client.create_workspace("My Team")

# Create issue
issue = await client.create_issue(
    workspace_id=workspace["id"],
    title="Bug Report",
    description="System crashes on startup"
)

Version Info

from praisonai_platform import __version__

print(f"Platform SDK version: {__version__}")
# Output: Platform SDK version: 0.1.0

---

Client Features

Authentication

Register, login, and manage JWT tokens

Workspaces

Multi-tenant workspace management

Issues

Issue tracking and project management

Agents

AI agent lifecycle management

---

Common Patterns

Server Deployment

from praisonai_platform import create_app

def deploy_server():
    """Deploy Platform API server"""
    app = create_app()
    
    # Add custom middleware if needed
    # app.add_middleware(...)
    
    return app

# For production
app = deploy_server()

SDK Integration

from praisonai_platform import PlatformClient

class AgentWorkflow:
    def __init__(self, platform_url: str, token: str):
        self.client = PlatformClient(platform_url, token)
    
    async def setup_workspace(self):
        """Create workspace and project"""
        workspace = await self.client.create_workspace("AI Project")
        project = await self.client.create_project(
            workspace["id"], 
            "Agent Tasks"
        )
        return workspace, project

Version Checking

from praisonai_platform import __version__
import sys

def check_compatibility():
    """Ensure compatible Platform SDK version"""
    required = "0.1.0"
    if __version__ < required:
        sys.exit(f"Platform SDK {required}+ required, got {__version__}")

---

Best Practices

Use Context Managers

Always use async context managers for HTTP clients to ensure proper cleanup:

async with PlatformClient("http://localhost:8000") as client:
    workspaces = await client.list_workspaces()
# Client automatically closes

Handle Authentication

Store and reuse JWT tokens for multiple requests:

client = PlatformClient("http://localhost:8000")
auth_data = await client.register("[e-mail]", "pass")
# Token is automatically stored in client._token

Error Handling

Wrap API calls in try-catch for HTTP errors:

try:
    workspace = await client.create_workspace("Duplicate Name")
except httpx.HTTPStatusError as e:
    if e.response.status_code == 409:
        print("Workspace name already exists")

Environment Configuration

Use environment variables for configuration:

import os
from praisonai_platform import PlatformClient

client = PlatformClient(
    base_url=os.getenv("PLATFORM_URL", "http://localhost:8000"),
    token=os.getenv("PLATFORM_TOKEN")
)

---

Related

Platform Architecture

Complete Platform SDK reference

Authentication Guide

JWT token management