Search

This is an example search page!

Reference:

• Talk to github repo using AI
• Leveraging AIChat for RAG with ILIMA THOUGHTS §

  Leveraging AIChat for RAG with a GitHub Repository

  This guide demonstrates how to use AIChat’s built-in vector database and full-text search capabilities to create a knowledge base from an entire GitHub repository. This enables Retrieval-Augmented Generation (RAG), allowing AIChat to provide more informed and contextually relevant responses based on the repository’s content.

  Creating a RAG Knowledge Base from a GitHub Repository

  Follow these steps to create a RAG knowledge base using AIChat:

  1. Initiate RAG: Use the following command within AIChat, replacing ilima-thoughts with your desired RAG name:

     > .rag ilima-thoughts
     

  2. Configure Embedding Model: AIChat will prompt you to select an embedding model. Choose the appropriate model based on your needs. For example:

     > Select embedding model: gemini:text-embedding-004 (max-tokens:2048; max-batch:100; price:0)
     

  3. Set Chunk Size and Overlay: Configure the chunk size and overlay for optimal retrieval performance. Experiment with these values to find what works best for your repository.

     > Set chunk size: 1500
     > Set chunk overlay: 100
     

  4. Add Documents: Specify the paths to the relevant directories within your GitHub repository. This example uses the unapologetic-thoughts repository. Adjust the paths to match your repository’s structure. Use the ** wildcard to include all files within the specified directories.

     > Add documents: https://github.com/igorlima/unapologetic-thoughts/tree/master/_posts/**; https://github.com/igorlima/unapologetic-thoughts/tree/master/snippets/**; https://github.com/igorlima/unapologetic-thoughts/tree/master/pages/**; https://github.com/igorlima/unapologetic-thoughts/tree/master/notebooks/**
     

  5. Exit RAG Configuration: Once you’ve added all the desired documents, exit the RAG configuration mode.

     @ilima-thoughts> .exit rag
     

  Managing Your RAG Knowledge Base

  AIChat provides several commands for managing your RAG knowledge base:

  • View Citation Sources: After a query, use the following command to see the sources used in the last query:

    .sources rag
    

    This helps you understand where AIChat is getting its information.

  • Edit Documents: Add or remove documents from an existing RAG knowledge base:

    .edit rag-docs
    

  • Rebuild RAG: After making changes to the documents in your repository, rebuild the RAG knowledge base to reflect those changes:

    .rebuild rag
    

  • Show RAG Information: View information about your RAG knowledge base, such as the number of documents and the embedding model used:

    .info rag
    

  Further Resources

  • AIChat RAG Guide: For a more in-depth explanation, refer to the official AIChat RAG Guide ⁺.
  • Leveraging AIChat RAG with Your RC File: Configuration Assistance Made Easy
  • Leveraging AIChat RAG with ILIMA SNIPPETS
  • Using RAG internally with the quick-notes markdown ⁺
    
    
• Fuse JS

  • FuseJS Home Page
  • Fuse Live Demo
  • Fuse options documentation
• the following vim hint is helpful when enabling the option to include the matches - includeMatches: true.
  • to move the cursor to a match position, use the one character in the search pattern /. and then move the N characters ahead by repeating the last search from the cursor’s position.
    • to go to the Nth match in the file: type the amount of movement to go ahead, then press the n - eg. 78n 250n.
    • the dot . matches any character except the new line.