@adrianfathurs/plugin-agent-tracker@1.0.3
NPM package to record AI prompting and dev activity into Mandays telemetry backend.
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe automatic global hook registration leaves a concrete agent integration risk, so this warrants a warning. The inspected telemetry sends identity and project metadata only when a token is configured, and does not include prompt contents.
Decision evidence
public snapshot- The postinstall script runs the setup code, which adds this package’s hooks to global Claude Code settings.
- The hooks are activated during setup and saved to the user’s Claude Code settings file.
- When Claude Code stops a turn, the hook collects the Git email, project name, prompt count, and duration for telemetry.
- Telemetry is sent only when an Agent Tracker token is configured, to the configured backend; the default endpoint is localhost.
- The inspected hook code records turn timing and project metadata; it does not send prompt text or tool contents.
- The package’s postinstall setup registers hooks but does not itself send telemetry.
Source & flagged code
5 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgThe postinstall script runs the setup code, which adds this package’s hooks to global Claude Code settings.
package.jsonView on unpkg · L12Findings
Affected versions and remediation
This report applies to @adrianfathurs/plugin-agent-tracker@1.0.3.
See all versions of @adrianfathurs/plugin-agent-tracker for other recorded verdicts.
- Review the evidence and your use of @adrianfathurs/plugin-agent-tracker@1.0.3 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.