AI Security Review
scanned 24m ago · by lpm-firewall-aiReview flagged AI-agent configuration or capability changes. This remains warn-only unless evidence shows foreign-agent hijack through preinstall/install/postinstall, hidden persistence, exfiltration, remote code execution, or other concrete malicious behavior.
Decision evidence
public snapshot- `dist/skill.js` installs/updates skills through `npx skills` and can replace `~/.claude/skills/pi-review`.
- `dist/update.js` explicitly runs `npm view` and `npm install -g @zephyrdeng/pi-review@latest`.
- `dist/panel-ui.js` launches a detached local dashboard process when `--ui web` is selected.
- `package.json` has `prepare` and `prepublishOnly`, but no `preinstall`, `install`, or `postinstall` hook.
- `dist/cli.js` reaches install, update, and uninstall paths only for explicit CLI commands.
- `dist/ui-server.js` binds the dashboard to loopback and requires a per-run capability token.
- No credential harvesting, exfiltration endpoint, dynamic evaluation, native loading, or destructive payload was found in inspected runtime sources.
Source & flagged code
4 flagged · loading sourcePackage source invokes a package manager install command at runtime.
dist/skill.jsView on unpkg · L14This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
dist/panel-cli-ui.test.jsView on unpkg