The application stores a seed commitment before its roll step, then reveals the seed afterward. You can check that the published proof is internally consistent. Because all roll inputs are server-chosen and there is no external pre-roll publication, this alone cannot prove the server never preselected a favorable input tuple.
Paste your pull ID below — usually shown on the pull result modal, or linked from the "View proof" button on any vault item.
See the actual rarity distribution across recent pulls vs the published tier weights — covers bounty pulls, packs, spin, and boxes.
View observed distribution →Revealed draws collected by the digest job are hashed into Merkle batches, and each batch links to the previous one. Cache today's chain tip and compare later to detect a rewrite after your snapshot.
View the chain →Plain-English explainer of the full stack — commit-reveal, Merkle digests, what's public vs private, and how to verify from the command line.
Read the explainer →Digest publish cadence, self-audit pass rate, open drift alerts, and the current chain tip — the transparency layer observing itself.
View health →