@atlasauth/embed@0.12.0
AI Security Review
scanned 11d ago · by lpm-firewall-aiDirect source inspection supports ordinary user-driven authentication and consent behavior, and the required runtime locations did not establish the reported account hijack or unrelated credential transfer. The scanner’s semantic parse limit applies to a virtual view and is not evidence of malicious behavior.
Decision evidence
public snapshot- The manifest has no install lifecycle hook and declares Atlas design and JavaScript SDK dependencies.
- Sign-up submits user-provided email and password to the Atlas client sign-up route; the SDK likewise starts sign-up using supplied values and consent.
- The inspected runtime helpers cover policy display and consent gating, sign-in attempts, email recovery, link polling, and SSO through the configured client.
- The widget renders sign-up and policy steps, and MFA enrollment uses the current client and account state.
- The inspected dependencies expose design and authentication SDK functionality and contain ordinary build scripts.
- The social-login bundle’s physical source was reviewed; the scanner’s virtual parse-limit label alone does not establish harmful behavior.
- Clicking the Atlas sign-in button starts an SSO popup flow for its configured connection.
Source & flagged code
7 flaggedBrowser source reuses an authenticated session to collect identity data and mutate account settings while reporting externally.
dist/embed.jsView on unpkg · L1The inspected runtime helpers cover policy display and consent gating, sign-in attempts, email recovery, link polling, and SSO through the configured client.
dist/embed.jsView on unpkg · L413Findings
Affected versions and remediation
This report applies to @atlasauth/embed@0.12.0.
See all versions of @atlasauth/embed for other recorded verdicts.
- Review the evidence and your use of @atlasauth/embed@0.12.0 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.