Verify a Random Draw
Paste a certificate code to check it.
Check the consistency of a draw certificate
Paste a supported certificate code to check whether its seed, settings, fingerprint and result agree. The verifier runs in your browser and does not upload the certificate.
The two checks
Fingerprint. The tool recomputes the SHA-256 fingerprint from the certificate’s seed and settings and compares it with the fingerprint stored in the certificate.
Replay. The tool reruns the deterministic generator using the seed and settings, then compares the computed sequence with the result in the certificate.
Both checks must pass. Success establishes the certificate’s internal consistency. A failed comparison shows a mismatch; it does not by itself establish who changed the data or when.
What a passing result cannot establish
This page cannot see an earlier public commitment or verify when it was published. Dates written inside the certificate are not independent evidence.
Someone can create a new certificate with a different seed, result and matching fingerprint, and that certificate can pass too. A seed can also be selected by trying alternatives until a preferred outcome appears. Publishing a commitment prevents changing that particular commitment afterwards, but does not prove the seed was selected impartially beforehand.
For independent scrutiny, you therefore need the original commitment published before the draw, an independently agreed way to choose the seed and the settings that define the draw. Compare the fingerprint shown here manually with the one in that earlier publication.
Run a draw others can check
- Agree on the participants or number mapping, settings and an independent method of selecting the seed.
- Create the fingerprint and publish it before the draw somewhere with an independent timestamp.
- Run the draw with the committed seed and settings.
- Publish the seed, result and certificate.
- Ask recipients to compare the certificate’s fingerprint with the earlier publication, then run the verification here.
The tool does not verify that these outside steps actually happened. They must be checked separately.
Why the settings matter
A seed is only one input. The algorithm, range, count, repetition rules and other options affect the output. The same seed with different settings can produce different numbers.
The fingerprint covers the seed and settings so that a recipient can compare them with a prior commitment. The verifier checks supported number-draw certificates from the random number generator; a seed from an unrelated website is not enough.