> ## Documentation Index
> Fetch the complete documentation index at: https://praison.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# launch

> Launch all agents as a single API endpoint (HTTP) or an MCP server.

Method

> This is a method of the [**Agents**](https://praison.ai/docs/sdk/reference/praisonaiagents/classes/Agents.md) class in the [**agents**](https://praison.ai/docs/sdk/reference/praisonaiagents/modules/agents.md) module.

Launch all agents as a single API endpoint (HTTP) or an MCP server.
In HTTP mode, the endpoint accepts a query and processes it through all agents in sequence.
In MCP mode, an MCP server is started, exposing a tool to run the agent workflow.

## Signature

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
def launch(path: str, port: int, host: str, debug: bool, protocol: str) -> Any
```

## Parameters

**path** `str` (default: '/agents')

API endpoint path (default: '/agents') for HTTP, or base path for MCP.

**port** `int` (default: 8000)

Server port (default: 8000)

**host** `str` (default: '127.0.0.1')

Server host (default: '127.0.0.1')

**debug** `bool` (default: False)

Enable debug mode for uvicorn (default: False)

**protocol** `str` (default: 'http')

"http" to launch as FastAPI, "mcp" to launch as MCP server.

### Returns

**Returns** `Any`

None


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.