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

# Generate Presentation over MCP

> Presenton includes an MCP server which can be used to generate presentation over AI chat and by agents


[Model Context Protocol](https://modelcontextprotocol.io/) is a standard that helps AI tools and models communicate and delete tasks to each other.

Presenton also supports presentation generation over MCP. Only `Streamable HTTP` is supported to connect with Presenton over the network, it does not support `stdio`.

For now one tool is available:

* **generate\_presentation**: Single endpoint to generate presentation with. `generate_presentation` accepts following fields:
  * content
  * instructions

You will need to connect to  `/mcp` for MCP server. If you're connecting the cloud then the endpoint is `https://api.presenton.ai/mcp`, and for local endpoints it should look something like `http://localhost:5000/mcp` based on your configuration.
