Are you tired of losing hours to endless scrolling on Twitter? FeedFocus simplifies your day by curating and summarizing the content that matters most to you. Stay updated on AI breakthroughs, industry trends, and niche discussions—all without the noise.
Reclaim your time: FeedFocus transforms hours of browsing into just minutes of actionable updates.
Follow these steps to set up FeedFocus on your machine:
Start by creating and activating a virtual environment:
python -m venv venv
source venv/bin/activate
Move to the project directory and install dependencies:
cd feedfocus
pip install -r requirements.txt
Remove ollama for the time being due to hallucinating results for 4bit models.
Download qwen2-7b-instruct-q4_k_m.gguf and place it in the data folder.
Run the Ollama server to host the Qwen model:
~~ ~~ollama pull qwen2:72b-instruct-q4_K_S ~~ ~~ollama serve ~~ ~~
- Copy the .env.example to .env and set GROQ_API_KEY
Update the configuration files: • config/models.py • config/platforms/twitter.py
Install the Tampermonkey extension for Chrome.
• Copy the script from userscripts/v3.js.
• Paste it into a new script in Tampermonkey and enable it.
Start the FeedFocus server with:
python twitter_server.py
Open x.com (formerly Twitter) and start scrolling. Use the Page Down key or scroll manually to let the script capture content for summarization.
Or run the below script to automatically press pagedown
python press_pgdown.py
