A Chrome extension that enables semantic search of your Twitter bookmarks using local LangChain embeddings. Find your bookmarked tweets by searching for concepts and meaning, not just keywords.
- Clone this repository
- Open Chrome and navigate to
chrome://extensions/ - Enable "Developer mode" in the top right corner
- Click "Load unpacked" and select the
extensiondirectory
- Semantic search across your Twitter bookmarks
- Uses LangChain and embeddings for concept-based matching
- Fast local search - no cloud/API dependencies
- Privacy-focused: all processing happens in your browser
- Find tweets based on meaning and concepts, not just exact text matches
- The extension processes your Twitter bookmarks locally
- Creates embeddings using LangChain to understand the semantic meaning of tweets
- When you search, it finds tweets that match the concept you're looking for
- Results are ranked by semantic similarity to your query
- Go to Twitter and open your bookmarks
- Use the search bar added by the extension
- Enter your search query - try searching for concepts!
- See results ranked by semantic relevance
manifest.json: Extension configurationcontent.js: Handles Twitter page integration and UIbackground.js: Manages bookmark processing and searchpopup.html/js/css: Extension popup interface
This project was heavily inspired by and built upon the functionality provided by the bookmarks-wrapped project by Sahil Lalani. Much of the core functionality, including bookmark processing, was adapted from his work. Special thanks to Sahil for laying the foundation.
# Load unpacked extension in Chrome for development
1. Enable Chrome Developer Mode
2. Load the /extension directoryThis extension:
- Runs 100% locally in your browser
- Does not send your bookmarks or searches to any server
- Does not require any API keys or cloud services
- Is completely open source
Contributions are welcome! Please feel free to submit a Pull Request.
MIT License - see LICENSE for details