Skip to main content
Blacklist
Sign in

Methodology

How Blacklist presents risk, where information comes from, and what results do and do not mean.

Risk bands

Public results show a band and a plain-language explanation. Only verified, strong evidence raises a band. Blacklist does not publish a numeric score.

  • Low riskNo significant risk indicators found. This is not a guarantee of safety.
  • Moderate riskSome risk indicators were found. Review the factors below before interacting.
  • High riskStrong risk indicators were found. Review the evidence carefully.
  • Severe riskVerified, serious risk indicators were found. Do not send funds without independent verification.
  • Not assessedThere was not enough usable data to assess this. This is not an indication of safety.

Disclosures

Technical capabilities of this address or contract. They are context only: common in legitimate projects and do not affect the risk assessment. Examples: mint or freeze authority, upgrade or pause authority, transfer hooks or fees, and source code that is not verified. They are listed separately and are never among the factors that explain a band.

Confidence

How complete and fresh the underlying data is. Independent of the band.

  • Low confidenceLimited or partial data was available.
  • Medium confidenceMost relevant data was available.
  • High confidenceComplete, recent data was available from verified sources.

Disputes and appeals

  • Disputed — under appeal. Filing an appeal never hides a finding; it is reviewed by someone who did not write or approve it.
  • Temporarily suspended pending review. In urgent cases a finding can be removed from assessments while it is reviewed.

How reviews work

  1. A signed-in member sends a private report, project suggestion or correction. Until staff review it, it has no public effect.
  2. A moderator reviews it and can ask the submitter for more information. Each decision is recorded in an audit history.
  3. Accepting a report drafts a finding. It is published only after approval, and a severe finding needs a second moderator who did not write it.
  4. Profile corrections and new project profiles are published by administrators, and the profile keeps its revision history.
  5. Anyone affected can appeal a published finding. Standing is checked per case, and the appeal is decided by someone who did not take part in the original decision.

Where information comes from

  • Source: Blockchain factDirectly observable blockchain data, shown with its finality and when it was observed.
  • Source: Blacklist analysisDeterministic, versioned analysis of blockchain facts. An interpretation, not a fact.
  • Source: Moderator findingReviewed and verified by Blacklist moderators against published criteria.
  • Source: Community allegationAn unverified claim submitted by a user. It has not been confirmed.
  • Source: External sourceThird-party data, such as source-code verification from a block explorer. Not Blacklist’s own finding.
  • Source: AI observationGenerated by a model. Never treated as fact.

Transfers and relationships

Interacting with a flagged address does not mean wrongdoing. Anyone can send funds to any address.

Strength

Observed
Seen directly in blockchain data.
Inferred
Derived by analysis. It may be wrong.
Verified
Confirmed by independent sources at the same snapshot.

Evidence

  • Seen in a transaction log
  • Reported by the asset contract’s own log
  • Confirmed by balance changes
  • Native transfer recorded by consensus

Authorization

  • Signed directly by the sending address.
  • Moved under an allowance or delegation granted by the sending address.
  • Executed by a protocol contract on behalf of the sending side.
  • Who authorized this transfer is unknown.