@zerohash-sdk/onboarding-js@1.15.13
A JavaScript SDK that enables frontend applications to seamlessly integrate with the Connect Onboarding product.
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe package’s default error path sends runtime telemetry containing participant-related JWT claims, browser user agent, and page origin to Zerohash collectors. It is limited to widget load failures and is not install-time behavior, but the package exposes no public opt-out, so this is suspicious telemetry rather than a clear attack.
Decision evidence
public snapshot- When widget loading fails, the loader calls its default reporter with error details, including the script URL and failure reason.
- The reporter sends a JSON POST to Zerohash Faro collector endpoints.
- The report includes decoded JWT participant and platform claims, plus the browser user agent and page origin.
- The report is triggered by the widget load path reached when the caller invokes render().
- The package has no install lifecycle scripts and exports its dist entry point.
- Reporting is limited to widget loading failures, rather than occurring on import or install.
- The collector endpoints belong to Zerohash, the onboarding SDK provider.
Source & flagged code
9 flaggedA bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
dist/index.d.tsView on unpkgWhen widget loading fails, the loader calls its default reporter with error details, including the script URL and failure reason.
dist/index.jsView on unpkg · L90Findings
Affected versions and remediation
This report applies to @zerohash-sdk/onboarding-js@1.15.13.
See all versions of @zerohash-sdk/onboarding-js for other recorded verdicts.
- Review the evidence and your use of @zerohash-sdk/onboarding-js@1.15.13 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.