@optimuslabs/optimus-discovery-staging
Shadow AI Discovery — daily heuristic scan for AI agents on macOS endpoints; writes a local JSON report.
This report shows the evidence for @optimuslabs/optimus-discovery-staging@0.4.8. 0 of 11 known versions are blocked or flagged. See all versions.
AI Security Review
scanned 13h ago · by lpm-firewall-aiThe CLI automatically uploads identifying machine details and AI discovery results to the vendor’s service, while the README describes the scan as local-only. This is suspicious runtime telemetry; the upload is tied to the package’s discovery feature and is not evidence of credential theft or unrelated collection.
Decision evidence
public snapshot- Running the CLI automatically starts a fire-and-forget upload after scanning.
- The upload payload includes the hardware UUID, hostname, username, scan summary, and discovered candidates and browser extensions.
- The report is POSTed to the configured endpoint; the default host is app.optimuslabs.io.
- The README says the scan has no network I/O and describes a local report, without disclosing the upload.
- The package exposes a CLI, so scanning and upload occur when it is run rather than on import.
- The uploaded discovery report is data produced by the package’s stated inventory feature and goes to the vendor’s service.
- The prepare script builds the package; no install-time data upload was identified.
Source & flagged code
5 flaggedThis package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
dist/discover.jsView on unpkgFindings
Affected versions and remediation
This report applies to @optimuslabs/optimus-discovery-staging@0.4.8.
The versions list below shows the verdict for each recorded version.
- Review the evidence and your use of @optimuslabs/optimus-discovery-staging@0.4.8 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.