Literature Guide

Daily automated digest of ML/AI research papers and opportunities, focused on mechanistic interpretability, AI safety, alignment, and related topics.

How It Works

A GitHub Actions workflow runs daily at 6am EST:

1. Fetches papers from arXiv, OpenAlex, Semantic Scholar, RSS blogs, Reddit, Bluesky, and newsletters
2. Deduplicates across sources using DOI matching and fuzzy title similarity
3. Ranks papers by relevance to target topics using Claude
4. Summarizes top 10-20 papers with key takeaways and related work
5. Stores each paper as structured markdown + JSON index
6. Generates a daily report in reports/
7. Notifies via Slack with highlights and a link to the full report

Setup

1. Create a GitHub repo and push this code
2. Add secrets in repo Settings > Secrets:
   • ANTHROPIC_API_KEY — from console.anthropic.com
   • SLACK_WEBHOOK_URL — from Slack app settings
   • SEMANTIC_SCHOLAR_API_KEY — from semanticscholar.org/product/api
3. The workflow runs automatically, or trigger manually from Actions tab

Local Development

pip install -r requirements.txt
export ANTHROPIC_API_KEY=sk-...
export SLACK_WEBHOOK_URL=https://hooks.slack.com/...
python src/main.py

Configuration

Edit config.yml to customize topics, sources, and thresholds.

Structure

• reports/ — Daily markdown reports
• papers/ — Per-paper markdown files with summaries
• papers.json — Structured index of all papers