Lenny RAG

Provides hierarchical RAG over Lenny Rachitsky podcast transcripts to surface insights, examples, and transcripts.

python

2

GitHub Stars

python

Language

4 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": {
    "mpnikhil-lenny-rag-mcp": {
      "command": "/path/to/lenny-rag-mcp/venv/bin/python",
      "args": [
        "-m",
        "src.server"
      ]
    }
  }
}

You run an MCP server that enables hierarchical retrieval over a large set of transcripts, surfacing targeted insights, real-world examples, and full transcript context. It provides a structured, scalable way to brainstorm product ideas by querying and drilling into topic-focused material from the Lenny RAG data set.

How to use

You interact with the Lenny RAG MCP Server through an MCP client. Start by launching the local server process and then connect your client to it using the provided stdio interface. Once connected, you can perform semantic search across the transcripts to discover relevant insights and examples, load complete topics with full context, view full episode transcripts, and browse available episodes with their expert tags.

Practical usage patterns include: - Semantic search to locate actionable insights or real-world examples related to a topic. - Drilling into a specific topic to view a topic summary, all insights, all examples, and the raw transcript segment. - Loading a complete transcript for an episode to review metadata and all topics present. - Listing episodes filtered by expertise to focus on areas like growth, pricing, or leadership.

How to install

# Prerequisites
- Python 3.8+ and a working Python environment
- Git for cloning the project

# Clone the repository (includes pre-built index via Git LFS)
git clone git@github.com:mpnikhil/lenny-rag-mcp.git
cd lenny-rag-mcp

# Create and activate virtual environment
python -m venv venv
source venv/bin/activate

# Install the package in editable mode
pip install -e .

# Claude integration (example for Claude CLI)
claude mcp add lenny --scope user -- /path/to/lenny-rag-mcp/venv/bin/python -m src.server

# JSON config example for Claude Desktop / Cursor / Claude JSON
{
  "mcpServers": {
    "lenny": {
      "type": "stdio",
      "command": "/path/to/lenny-rag-mcp/venv/bin/python",
      "args": ["-m", "src.server"],
      "cwd": "/path/to/lenny-rag-mcp"
    }
  }
}

Additional notes

Replace /path/to/lenny-rag-mcp with your actual clone location in all configurations. The server runs locally via a Python process and communicates through the standard input/output interface configured by your MCP client.

Project structure and goals

The server provides a hierarchical extraction and retrieval flow over 299 transcripts. It breaks down each episode into topics, insights, and examples to enable progressive disclosure and efficient context management during searches.

Available tools

search_lenny

Semantic search across the entire corpus returning pointers for progressive disclosure with episode references and topic IDs.

get_chapter

Load a specific topic with full context, including topic summary, all insights, all examples, and the raw transcript segment.

get_full_transcript

Load the complete episode transcript with metadata, enabling full review of the episode and its associated topics.

list_episodes

Browse episodes with optional expertise filtering to surface guest names and tags.