Regulatory & legal standing · 25%
Verify the operating entity, official licence, permitted activities, regional authorisation and unresolved regulatory action.
Cexvia Exchange Risk Rating / V1.0.0
Cexvia Exchange Risk Rating V1 uses five weighted dimensions. Higher scores indicate lower observed risk under current public evidence; redlines, coverage and confidence are applied separately.
Verify the operating entity, official licence, permitted activities, regional authorisation and unresolved regulatory action.
Assess legal entities, ownership relationships, named leadership, contracting terms, addresses and material entity changes.
Review proof of reserves, update frequency, verified wallets, user inclusion, asset composition, liabilities and assurance scope.
Track major attacks, security controls, certifications, reimbursement, disclosure quality and independently verifiable remediation.
Combine funding-channel stability, official status history, service checks, complaint and appeal mechanisms, verified complaints and legal continuity.
Apply serious-event redlines after the weighted score, then enforce the confidence cap and publish qualitative exposure guidance.
01 Collect official status, probes, incident archives, reserve disclosures, controls, legal records, domain changes, vulnerability discovery and verified complaints.
02 Resolve every observation to the affected exchange product, legal entity, region and time window.
03 Validate source quality, materiality, duplication, loss amounts, remediation and current status. Automated output remains unreviewed until this gate passes.
04 Publish an alert or rating input only after review. Reachability, CVE keywords, anonymous posts and complaint volume never become direct risk facts by themselves.
↑ Scores run from 0 to 100; higher means lower observed risk under public evidence
25 Regulatory 25% · operations and user protection 25%
20 Assets 20% · corporate transparency 15% · security 15%
! Critical redlines cannot be cancelled out by strong scores elsewhere
∅ Missing data lowers coverage or confidence; it is never guessed or automatically scored as zero