AI Security Review
scanned 3h ago · by lpm-firewall-aiNo confirmed malicious attack surface is established. Risky primitives are present but tied to explicit deployment, scaffolding, adoption, tunnel, or demo commands.
Decision evidence
public snapshot- dist/demo.js can run Docker, SSH, git-like CLI flows and Cloudflare tunnel deletion/creation when directly invoked with up/down/clear.
- dist/init/init.js runs git init and pnpm install during explicit intentic init scaffolding.
- dist/adopt/adopt.js commits/pushes local intent repos to user Forgejo using provided credentials.
- dist/resolve/resolve.js dynamically imports the user-supplied deploy config path for resolution.
- package.json has no preinstall/install/postinstall lifecycle hooks; bin only points to dist/cli.js.
- dist/cli.js only dispatches the Stricli command app; no install-time or import-time payload seen in CLI entrypoint.
- Network use is deployment-aligned: Cloudflare API, user domains, local demo services, Forgejo control-plane operations.
- Secrets are read from user env/.env or generated stores for deployment workflows; no evidence of covert credential harvesting or third-party exfiltration.
- Filesystem writes are scoped to explicit CLI outputs such as intent/, desired-state/, app/, .secrets.json, access/status files, or demo state.
- No AI-agent control-surface mutation, persistence hook, destructive broad filesystem behavior, or remote payload execution found.
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