Server: Agents MCP

Deploy agents as MCP servers using Agent.launch(protocol=‘mcp’)

Deploy single or multi-agent systems as MCP (Model Context Protocol) servers for Claude Desktop, Cursor, and other MCP clients.

Quick Start

1. Install Dependencies

   pip install "praisonaiagents[mcp]"
   
2. Set API Key

   export OPENAI_API_KEY="your-key"
   
3. Create Agent MCP Server

   from praisonaiagents import Agent
   
   agent = Agent(
       name="TweetAgent",
       instructions="Create engaging tweets",
       llm="gpt-4o-mini"
   )
   agent.launch(port=8080, protocol="mcp")
   
4. Verify

   curl http://localhost:8080/sse
   

Single Agent as MCP

from praisonaiagents import Agent

agent = Agent(
    name="TweetAgent",
    instructions="Create engaging tweets",
    llm="gpt-4o-mini"
)
agent.launch(port=8080, protocol="mcp")

Expected Output:

🚀 Agent 'TweetAgent' MCP server starting on http://0.0.0.0:8080
📡 MCP SSE endpoint available at /sse
📢 MCP messages post to /messages/
🛠️ Available MCP tools: execute_tweetagent_task

Multi-Agent as MCP

from praisonaiagents import Agent, AgentTeam

researcher = Agent(name="Researcher", instructions="Research topics", llm="gpt-4o-mini")
writer = Agent(name="Writer", instructions="Write content", llm="gpt-4o-mini")

agents = AgentTeam(agents=[researcher, writer])
agents.launch(port=8080, protocol="mcp")

Expected Output:

🚀 Agents MCP Workflow server starting on http://0.0.0.0:8080
📡 MCP SSE endpoint available at /sse
📢 MCP messages post to /messages/
🛠️ Available MCP tools: execute_workflow
🔄 Agents in MCP workflow: Researcher, Writer

MCP Endpoints

Endpoint	Method	Description
/sse	GET	SSE connection for MCP
/messages/	POST	Send MCP messages

launch() Parameters

Parameter	Type	Default	Description
port	int	8000	Server port
host	str	0.0.0.0	Server host
protocol	str	mcp	Must be mcp
debug	bool	False	Debug mode

Connect MCP Client

Claude Desktop (claude_desktop_config.json):

{
  "mcpServers": {
    "praisonai-agent": {
      "url": "http://localhost:8080/sse"
    }
  }
}

Cursor (.cursor/mcp.json):

{
  "mcpServers": {
    "praisonai-agent": {
      "url": "http://localhost:8080/sse"
    }
  }
}

Test MCP Tools

# List tools
curl -X POST http://localhost:8080/messages/ \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc": "2.0", "method": "tools/list", "id": 1}'

# Call tool
curl -X POST http://localhost:8080/messages/ \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "method": "tools/call",
    "params": {"name": "execute_tweetagent_task", "arguments": {"task": "Create a tweet about AI"}},
    "id": 2
  }'

Docker Deployment

app.py:

from praisonaiagents import Agent

agent = Agent(instructions="Create tweets", llm="gpt-4o-mini")
agent.launch(port=8080, protocol="mcp")

Dockerfile:

FROM python:3.11-slim
WORKDIR /app
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
COPY app.py .
EXPOSE 8080
CMD ["python", "app.py"]

requirements.txt:

praisonaiagents[mcp]

Build and run:

docker build -t agent-mcp-server .
docker run -p 8080:8080 -e OPENAI_API_KEY=$OPENAI_API_KEY agent-mcp-server

Troubleshooting

Issue	Fix
Port in use	lsof -i :8080
Missing deps	pip install "praisonaiagents[mcp]"
No API key	export OPENAI_API_KEY="your-key"
SSE not connecting	Check firewall, use host="0.0.0.0"

Related

• Tools MCP - Deploy tools as MCP server
• Recipes MCP - Deploy recipes as MCP server
• PraisonAI MCP - Full PraisonAI MCP server
• Agents HTTP - Deploy as HTTP server