chronicles-of-aethelgard

Use Rag

This sample is using qwen2.5:1.5b for the chat completion and snowflake-arctic-embed:33m for the embeddings generation. Don't forget to pull the latest version of the models before running the script:

ollama pull qwen2.5:1.5b
ollama pull snowflake-arctic-embed:33m

The settings for Bob are stored in .bob directory.

You have to create a rag.json file in the .bob directory with the following content:

{
    "chunkSize": 1024,
    "chunkOverlap": 256,
    "similarityThreshold": 0.3,
    "maxSimilarity": 10
}

Creation (or update) of the Vector store

Run the following command in the current directory to create or update the Vector store using the content in the content directory:

bob --rag ./content

Only text files are considered for the Vector store (md, txt, etc.)

Completion

Run the following command in the current directory to generate a report using the prompt in prompt.md:

bob 

#or
bob --user "Summarize information on Gnomish Species"

Bob will detect automatically that's a RAG project