Grade corrections & disputes

What a CheckCrypto grade is

• Our grades (A–F) are a risk estimate (uncalibrated) produced by automated analysis of public on-chain data and third-party security signals.
• A grade is not a recommendation or advice to buy, sell, or hold any crypto asset, and not a conclusive finding that the issuer has broken the law.
• Every grade is shown together with the evidence signals used to produce it, and the verification links on the page let you check those facts yourself.
• Grades can change at any time as new data is collected.

When to file a dispute

• False positive — a legitimate project shown with a high risk level
• False negative — a confirmed scam shown with a low risk level
• Factual error — a stated basis (honeypot, tax, liquidity, etc.) differs from on-chain fact (e.g. ownership already renounced but shown as not renounced)
• Stale data — steps taken by the project (liquidity lock, ownership renouncement, completed audit) are not yet reflected in the grade
• Misidentification / impersonation — another token's data was linked by mistake, or a legitimate contract was flagged as impersonation (or the reverse)

An opinion on its own — "the assessment is unfair" — is not grounds for a correction. Corrections are made when there is verifiable factual evidence.

Three required elements — we cannot review without all three

1. The incorrect data point — which token (chain + contract address required) and which signal / score / evidence item is at issue
2. The correct value — what it should actually be
3. A publicly verifiable evidence link — an on-chain transaction, official announcement, audit report, source code, or any source anyone can check

Reports without an evidence link are not reviewed. Claims alone cannot change a verdict — the same standard by which our verdicts are tied to reproducible facts.

Telling us whether you are affiliated with the project helps our review (it counts against you in no way). A contact for a reply is optional.

How a dispute is handled

1. Requests are handled in regular batches — they are not processed the moment they arrive, and we generally do not send individual replies.
2. Review — we cross-check the submitted evidence against on-chain data and re-run the assessment.
3. Outcome — (a) corrected and re-scored, (b) rejected for insufficient evidence, or (c) a request for further material.
4. Every request is read, and we judge on reproducible facts (on-chain data, simulation). When a correction is made, the re-assessment is reflected on that token's page, where anyone can see it.

Principles, and the record of corrections

• When an error is confirmed we re-run the assessment to correct it and improve the methodology where needed. Correction history is kept as an internal record, noting that the original grade was produced from the data available at that time.
• The score is an automated estimate, not a value judgment — some signals (e.g. mintable, ownership not renounced) can be normal by a project's design, and we reflect such cases when you tell us.
• However, reproducible hard facts (honeypot simulation, etc.) are not removed on an unsubstantiated request alone.

If you're curious how we assess, see the Methodology; for terms, see the Glossary.