starter-cleaner

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes multiple project-level commands including pnpm install, pnpm type-check, pnpm lint, and pnpm dev:h5. These are used to update the lockfile, verify code integrity, and trigger configuration generation as part of the template cleaning process.
  • [COMMAND_EXECUTION]: The provided utility script (scripts/clean.js) performs destructive file system operations using fs.rmSync to delete directories such as docs/, src/subPages/, and src/subEcharts/. This behavior is the primary stated purpose of the skill and is constrained to a hardcoded list of relative paths.
  • [SAFE]: The skill includes several security and safety best practices: it requires explicit user confirmation for destructive actions, provides a --dry-run flag to preview deletions without executing them, and implements a project identity check that verifies the package.json name and version before proceeding.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 02:34 PM
Security Audit — agent-trust-hub — starter-cleaner