AI Security Review
scanned 3h ago · by lpm-firewall-aiNo confirmed malicious attack surface. The package provides explicit user-invoked game scaffolding, local CLI utilities, and a Workshop MCP server for the current project.
Decision evidence
public snapshot- `package.json` has only `prepublishOnly`; no install-time lifecycle hook.
- `bin/create-hayao.mjs` only scaffolds files into the explicitly named project directory.
- `bin/hayao-mcp-cli.mjs` starts the packaged Workshop MCP server only when the user invokes `hayao-mcp`.
- `dist/mcp.js` confines workshop reads/writes to the current project's `.workshop/` data and exposes user-invoked local tooling.
- `dist/mcp.js` dynamic import loads discovered local game modules; `new Function` is bundled AJV validation code, not package payload execution.
- No shipped executable contains outbound HTTP/fetch/WebSocket calls or credential/environment harvesting.
Source & flagged code
5 flagged · loading sourcePackage source references child process execution.
bin/hayao-mcp-cli.mjsView on unpkg · L5Package source invokes a package manager install command at runtime.
bin/hayao-mcp-cli.mjsView on unpkg · L9Package source references dynamic require/import behavior.
dist/mcp.jsView on unpkg · L6Package source references a known benign dynamic code generation pattern.
dist/mcp.jsView on unpkg · L2944This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
bin/hayao.mjsView on unpkg