Turn your chat history into a searchable social memory database
T

Turn your chat history into a searchable social memory database

Turn your chat history into a searchable social memory database

6,827 stars
N/A forks
N/A contributors

README

Project documentation from GitHub

Turn Your Chat History into a Searchable Social Memory Database

Ever wish you could actually find that brilliant idea, useful link, or key decision buried somewhere in your endless Slack, Discord, or Telegram history? We've all been there. Our chat apps are great for real-time conversation but terrible for long-term knowledge retrieval. What if you could treat your collective chat history like a personal, searchable database?

That's the idea behind ChatLab, a clever open-source project that transforms your exported chat logs into a semantic search engine. It's like giving your team's forgotten conversations a second life.

What It Does

ChatLab is a Python-based tool that takes your exported chat history (currently supporting Slack, Discord, and Telegram) and processes it through a local embedding model. It converts your conversations into vector embeddings and stores them in a local SQLite database with ChromaDB. The result is a private, offline search interface where you can ask questions in natural language and get back relevant snippets from past discussions, complete with speaker names and timestamps.

Why It's Cool

The beauty of ChatLab is in its practical, privacy-focused approach. Unlike cloud-based services, everything runs locally on your machine—your data never leaves your computer. It uses the lightweight all-MiniLM-L6-v2 model by default, so you don't need a beefy GPU to get started.

It's also surprisingly simple. You export your data from your chat platform (usually a JSON or ZIP file), point ChatLab at it, and let it build the index. The search interface is a straightforward local web app. You can ask things like "What did we decide about the API design last month?" and actually get answers, pulling context from across different channels and time periods.

For developers, it's a great example of a usable RAG (Retrieval-Augmented Generation) pipeline built with common open-source tools: ChromaDB for the vector store, Sentence Transformers for embeddings, and FastAPI for the web layer. The code is clean and easy to follow if you want to extend it.

How to Try It

Getting started is pretty straightforward if you have Python experience:

  1. Clone the repo:

    git clone https://github.com/hellodigua/ChatLab.git
    cd ChatLab
    
  2. Set up a virtual environment and install:

    python -m venv venv
    source venv/bin/activate # On Windows: venv\Scripts\activate
    pip install -r requirements.txt
    
  3. Export your data from Slack, Discord, or Telegram (check the repo's README for platform-specific export guides).

Did you like this issue?

Join our weekly newsletter

Love discovering amazing projects?

Help us continue bringing you the best open-source discoveries every week.

Back to Projects
Last updated: Mar 8, 2026