ChatPPT

Provides multi-document PPT indexing, RAG-based Q&A, and MCP tool integration for interactive analysis.

python

13

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "pingcy-app_chatppt": {
      "command": "python",
      "args": [
        "mcp_ppt_server.py",
        "--transport",
        "sse"
      ]
    }
  }
}

ChatPPT MCP Server enables a multi-document RAG engine for PPTs with MCP integration, allowing you to perform smart question‑answering and analysis across multiple presentations directly through MCP-enabled clients.

How to use

You interact with the ChatPPT MCP Server through an MCP client. Start the MCP server in stdio mode to run locally, then connect your client to it to perform intelligent Q&A and analysis over indexed PPT documents. Use the interactive test tools to verify capabilities and experiment with different prompts and document sets. The MCP flow lets your client call the server’s tools to fetch, reason about, and summarize content from multiple PPT documents.

How to install

Prerequisites: Python 3.8 or newer, and a working Python package manager (pip). Ensure you have network access if you plan to fetch dependencies from PyPI.

Install the project in editable mode to install local dependencies and set up the package.

pip install -e .

Prepare environment variables by copying the example template and customizing values as needed.

cp .env.example .env

Run the interactive test interfaces to verify RAG and MCP functionality.

python rag_interactive_test.py

Start the MCP server in stdio mode (this initializes the MCP endpoint you will connect to from the client). The example shows starting the MCP server with SSE transport. Then you can run the MCP interaction tests.

python mcp_ppt_server.py --transport sse

python mcp_interactive_test.py

Additional notes

Project structure includes core components such as the RAG engine and document processor. The RAG engine is responsible for indexing and retrieving content from multiple PPT documents, while the document processor handles PPT processing and integration with vector embeddings.

Configuration and startup details

You can launch the MCP server in stdio mode to enable client connections via MCP. The command shown is used to start the server in SSE transport mode, which forwards MCP messages over a streaming interface.