Summary

Add PraisonAI to the "📚 Frameworks → For clients" section as an AI Agents framework with native MCP client support.

About PraisonAI

• GitHub: https://github.com/MervinPraison/PraisonAI
• PyPI: https://pypi.org/project/praisonai/ (2M+ downloads)
• MCP Registry: https://registry.modelcontextprotocol.io/servers/io.github.MervinPraison/praisonai
• Documentation: https://docs.praison.ai/mcp

MCP Integration

PraisonAI provides a native MCP class that allows AI agents to consume any MCP server as tools:

from praisonaiagents import Agent
from praisonaiagents.mcp import MCP

agent = Agent(
    instructions="You are a helpful assistant",
    tools=MCP("npx -y @modelcontextprotocol/server-filesystem /tmp")
)
agent.start("List files in /tmp")

Supported MCP Features

• All transports: stdio, SSE, WebSocket, Streamable HTTP
• Environment variables: Pass API keys and config to MCP servers
• Multiple servers: Use multiple MCP servers simultaneously
• Async support: Full async/await compatibility

Why This Belongs in Frameworks (For clients)

1. Client-side MCP integration - PraisonAI is an MCP client that consumes MCP servers
2. Framework, not a server - It's a framework for building AI agents, not an MCP server implementation
3. Already in MCP Registry - Published at io.github.MervinPraison/praisonai
4. Active maintenance - Regular releases, 10K+ GitHub stars, 2M+ PyPI downloads