AI Security Review
scanned 4h ago · by lpm-firewall-aiLPM blocks this version under the AI-agent control-surface policy. On npm installation, the postinstall hook invokes `setup-skills`. The bundled CLI writes or symlinks Shopport skills into multiple AI-agent home-directory skill locations, including `~/.claude/skills` and `~/.codex/skills`, without a separate user command.
Decision evidence
public snapshot- `package.json` runs `dist/cli.js setup-skills` in `postinstall`.
- `dist/cli.js` enumerates home-directory skill paths for Claude Code, Codex, Gemini CLI, OpenCode, and other agents.
- The installer creates each target skills directory and symlinks/copies bundled skills there.
- It removes prior package-namespaced skill entries from every enumerated target.
- `dist/skills/README.md` confirms automatic registration during npm installation.
- Observed installer targets only packaged `dist/skills/*` directories.
- No network endpoint is used by the inspected setup-skills installation path.
- No credential harvesting or exfiltration was established in the install path.
- Uninstall removes only shopport-namespaced skill entries.
Source & flagged code
5 flagged · loading sourceInstall-time lifecycle script matches a deterministic static-gate block pattern.
package.jsonView on unpkgPackage defines install-time lifecycle scripts.
package.jsonView on unpkgPackage source references a known benign dynamic code generation pattern.
dist/dashboard-ui/assets/index-ChYTl53S.jsView on unpkg · L467Package contains source files above the static scanner size ceiling.
dist/cli.jsView on unpkgPackage contains an oversized executable-looking CLI entrypoint.
dist/cli.jsView on unpkg