AI Security Review
scanned 2h ago · by lpm-firewall-aiNo confirmed malicious attack surface was established. The package is a deployment CLI whose shell, filesystem, SSH, and network behavior is exposed through explicit user-invoked commands and is consistent with its intent-driven deployment purpose.
Decision evidence
public snapshot- dist/demo.js combines docker/git execution, Cloudflare API calls, SSH setup, and local secret writes for explicit demo up/clear modes
- dist/init/init.js and dist/init/scaffold-app.js run git/pnpm and write scaffold files when user invokes init
- dist/resolve/resolve.js dynamically imports the user-supplied config path and reads a Cloudflare token from env for zone discovery
- dist/apply/apply.command.js can POST a reconcile summary to a graph-provided Discord webhook
- package.json has no preinstall/install/postinstall lifecycle hooks
- dist/cli.js only invokes @stricli routing for user-supplied CLI commands
- Network calls are aligned with deployment features: Cloudflare, Forgejo/control plane, local demo services, and optional Discord webhook
- No import-time credential harvesting, hidden persistence, foreign AI-agent config mutation, or stealth payload download observed
- Command execution is tied to explicit user commands such as init, adopt, apply, and demo, not automatic install execution
Source & flagged code
4 flagged · loading sourcePackage source references child process execution.
dist/init/scaffold-app.jsView on unpkg · L1Package source references dynamic require/import behavior.
dist/resolve/resolve.jsView on unpkg · L6A single source file combines environment access, network access, and code or shell execution; review context before blocking.
dist/demo.jsView on unpkg · L29Source combines command execution, command-output handling, and outbound requests; review data flow before blocking.
dist/demo.jsView on unpkg · L29