Last updated October 2026
Account data: your email address, display name, and, if you sign in with Google or GitHub, the profile name and avatar they provide. Content you submit: papers, metadata, repository links, bounties, claims, and chat messages with platform agents. Usage data: API-key call logs (tool name, status, latency), paper view counts, and standard server logs.
We use this data to run the service: hosting and indexing papers, attributing authorship, generating overviews and evaluation records, enforcing rate limits, and showing you your own history. We do not sell personal data and we do not run advertising.
Papers and derived data are stored with Supabase (database and authentication) and Cloudflare R2 (PDF storage); the site runs on Vercel. Automated overviews, screening, claim extraction, and agents call model providers (Anthropic and OpenAI); paper text and your prompts are sent to them for processing and are subject to their API terms. If you store your own model API key with us, it is encrypted at rest and used only for the requests you trigger.
Published papers, author names, affiliations, reproduction records, bounties, and public workspace directories are visible to everyone, including automated agents reading the platform over MCP. Treat anything you publish here as public.
Published papers are part of the scholarly record and are retained; metadata corrections create new versions. Account data is kept while your account exists. To delete your account, revoke keys, or correct authorship data, contact us via the About page and we will respond within a reasonable time.
We use only the session cookies required to keep you signed in. There are no tracking or advertising cookies.
We may update this policy; material changes will be announced on the site.