What problem does it solve? Researchers need a way to browse Paperzilla project feeds, read canonical papers, and manage recommendation feedback directly from an AI agent without switching to a browser or writing API calls by hand. ## Core Features & Use Cases - Project and Feed Browsing: List projects and browse recommendation feeds with filters like --must-read, --since, and --limit, plus JSON and Atom output. - Paper and Recommendation Access: Fetch canonical papers and project recommendations as JSON or markdown for summarization and analysis. - Feedback Management: Upvote, downvote, or star recommendations and export feeds for external readers. - Use Case: Ask your agent for the latest must-read recommendations from a project since a given date, then have it open one recommendation, fetch the paper as markdown, and summarize why it matters to your research. ## Quick Start Ask the agent to list your Paperzilla projects and show the latest must-read recommendations from one of them.