@skillsmith/cli@0.8.12
A registry for sharing, scanning, and tracking agent skills across teams.
AI Security Review
scanned 5d ago · by lpm-firewall-aiThe package’s install lifecycle does not run the flagged behavior. The telemetry hook is installed and enabled through separate CLI commands, and its collection is gated by local opt-in configuration.
Decision evidence
public snapshot- The manifest exposes a CLI and has only a prepublishOnly script for a publish guard, build, and tests; there is no install lifecycle hook.
- Telemetry hook installation is a separate CLI command, and it instructs the user to run `skillsmith telemetry enable` before collection starts.
- The hook exits unless local configuration sets telemetry.enabled to true; it reads only the session ID and skill name from the hook payload, not tool arguments.
- When enabled, the hook sends an anonymous skill-invocation event with the skill name, session ID, duration, and platform; the script states that paths and environment variables are excluded.
- The telemetry template contains an outbound endpoint, but sending is gated by explicit telemetry enablement.
- The CLI writes a hook to Claude settings only through the explicit install-hook command; this does not happen during package installation.
Source & flagged code
12 flaggedSource appears to send environment or credential material to an external endpoint.
dist/cli.jsView on unpkg · L50Findings
Affected versions and remediation
This report applies to @skillsmith/cli@0.8.12.
See all versions of @skillsmith/cli for other recorded verdicts.
- Review the evidence and your use of @skillsmith/cli@0.8.12 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.