The Agent Skills Directory

[SAFE]: The skill is entirely composed of documentation and instructional markdown files designed to improve code quality during human or agent reviews.
[SAFE]: No sensitive data exposure, hardcoded credentials, or exfiltration vectors were found. Any mentions of external APIs (Stripe, Shopify) are used purely as examples in the documentation.
[SAFE]: Command-line heuristics provided in the 'Detection' sections of the rules are standard utilities (grep, find, eslint, phpstan) used for static code analysis and are safe for use in a developer environment.
[SAFE]: There are no signs of obfuscation, multi-layer encoding, homoglyphs, or hidden URLs within the skill's files.
[SAFE]: All external references, such as arXiv research papers and community-standard tools (ESLint, PHPStan, Zod), are legitimate industry resources.
[SAFE]: The skill does not contain any privilege escalation, persistence mechanisms, or dynamic execution patterns.

code-slop