Procurement & security summary.
What a security, privacy or procurement reviewer asks first about CREdocket, gathered in one place, each point linked to the full page it comes from. No pricing or plan terms are on this page — that's a separate conversation with us directly.
A commercial real estate litigation intelligence product: a research-driven tracker of CRE lawsuits and bankruptcies, a portfolio monitor that checks a subscriber's saved companies against new public filings, and a pair of beta AI-assisted tools. It is run as a small, Supabase-hosted operation — our Data Security & Governance page says so directly rather than implying a larger security team than actually exists, and states plainly where we don't yet have a firm answer to a question.
- Account data (email and password) is handled by Supabase, our backend provider, and used to sign you in and associate your Watchlists, Portfolio entities, purchases and usage limits with your account.
- For the Case Value Calculator and the Lease Clause Redline Checker, the raw file you upload is never sent to any server: text extraction happens entirely in your own browser (PDF via pdf.js, .docx via mammoth.js, .txt via the browser's own file reader).
- The extracted text is then sent to Anthropic's commercial API (Claude) to generate the analysis. Nothing in our integration opts into any Anthropic feature that trains a model on your content.
- Watchlist and Portfolio entity names (a property, tenant, lender, guarantor or counterparty you add) are stored to run the daily match against new filings, described in full on the Data Security page.
Full detail, including what we have and haven't independently verified: Data Security & Governance.
We have not negotiated a Zero Data Retention agreement with Anthropic. Anthropic's published policy for a standard commercial API customer — which is what we are — states that inputs and outputs are automatically deleted from their backend within 30 days of receipt or generation, with narrow exceptions for a legal hold or a subsequent negotiated retention arrangement.
Separately, content flagged by Anthropic's own automated trust-and-safety systems can be retained longer, up to two years, regardless of that retention policy — an exception on Anthropic's side, not ours, that would only apply if content triggered their abuse detection. Sources for both figures are cited directly on the Data Security page.
You may cancel a subscription at any time. Canceling stops future billing; it doesn't retroactively delete past transaction records. See our Privacy Policy for the full statement, and our Terms of Service for the governing account terms.
Two pages are kept current automatically, straight from the live data, so the numbers here can't go stale the way a copied-in figure would: Coverage states exactly what the tracker sees — how many matters, federal versus state, docket-linked versus sourced to reporting, and where coverage is dense or thin by state and practice area — and what Portfolio monitoring covers and doesn't (every Chapter 11 petition nationwide and matching federal suits; two named state court systems; nothing else).
The Alert Log is rewritten by the monitoring job itself after every run: every surveillance run in the last 30 days, which ones failed and why, and the median time from a match to the alert email, so reliability can be checked rather than taken on faith.
Privacy Policy · Terms of Service · Litigation Tools Terms (the AI-assisted tools specifically) · Methodology · Corrections log
Questions a reviewer needs answered that aren't covered above: contact us directly.