@connect-xyz/withdraw-js@0.54.11
A JavaScript SDK that enables frontend applications to seamlessly integrate with the Connect Withdraw product.
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe SDK automatically sends runtime failure telemetry containing browser and application context and JWT-derived claims to its vendor’s collector, with no documented opt-out in the inspected public API. This is a warning-level telemetry concern; the report is limited to widget-load failures and does not include the raw JWT.
Decision evidence
public snapshot- The loader reports through its callback when loading the widget script fails.
- The report payload includes JWT-derived partner and participant claims, browser user-agent, and page origin.
- It sends the report as a JSON POST with a Faro session ID header.
- The configured collector endpoints are Zerohash Grafana Faro collectors.
- A generated type comment mentions failure reporting to Faro with JWT claims, but the inspected public API shows no telemetry opt-out.
- The report runs on widget script-load failure, after the application calls render().
- The code decodes selected JWT claims; it does not send the raw JWT in the report.
- The package manifest has no lifecycle scripts.
Source & flagged code
7 flaggedA bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
dist/index.d.tsView on unpkgA generated type comment mentions failure reporting to Faro with JWT claims, but the inspected public API shows no telemetry opt-out.
dist/index.d.tsView on unpkg · L227Findings
Affected versions and remediation
This report applies to @connect-xyz/withdraw-js@0.54.11.
See all versions of @connect-xyz/withdraw-js for other recorded verdicts.
- Review the evidence and your use of @connect-xyz/withdraw-js@0.54.11 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.