: The core workflow is simple: search for a paper or paste its identifier, and the tool returns a structured summary with key points, results and limits. If a paper has been summarized before, it opens immediately; if not, the AI reads and summarizes it, showing progress while you wait. On the paper page you can open the PDF beside the summary, request another language or reading level, and start a conversation grounded in the paper.
What sets it apart is its emphasis on traceability. The chat assistant answers only from the paper's text, links each claim to the supporting sentence and marks numbers or statements that no paper states. For paywalled papers or restrictive licenses, it says plainly that the summary uses the abstract rather than the full text, so you always know what the AI actually read.
Around the summaries sits a full research workflow: collections with note-taking and comparison tables, a daily paper alert and weekly field letters, bibliography checks, rising-topic tracking and integrations with tools like Zotero, Overleaf, Notion and Obsidian. It is designed for students, researchers and anyone who wants to keep up with a scientific field without reading every paper cover to cover.
