browser-assistant

An AI browser assistant addon for Firefox that can read the text on the screen and have a chat about it. Tested with LLM backend Ollama assumed to be hosted at http://localhost:11434.

TODO

• Talk to Ollama
• Send text on screen as part of input context
• Use vision model to dynamically generate enriched alt text for images to be sent as text to the input context.
• Make model and endpoint configurable
• Use langchain to chunk and embed the page for real-time RAG

Loading the extension in Firefox

Load the Extension in Firefox:

1. Install Firefox Developer Edition
2. Open Firefox and navigate to about:debugging
3. Click on "This Firefox" on the left hand side and then "Load Temporary Add-on".
4. Select the manifest.json file from your extension directory.

Ollama

Be sure to set OLLAMA_ORIGINS=* environment variable. See "how do I configure Ollama server?"