The Agent Skills Directory

[SAFE]: The skill's core instructions and reference documents promote robust security practices, specifically advising the agent to handle external input as untrusted (unknown) and to enforce identity/tenant authorization at trusted boundaries.
[SAFE]: No obfuscated content, Base64-encoded commands, homoglyph attacks, or zero-width character exploitation was detected across the 294 analyzed files.
[SAFE]: The project's dependencies are limited to the official TypeScript compiler (typescript@5.9.3), and all external source references point to well-known documentation (e.g., official TypeScript, Node.js, OWASP, and Playwright repositories).
[SAFE]: The utility scripts (install-skill.mjs, package-plugins.mjs, and validate.mjs) are standard development tools used for local file copying and verification. They do not contain any network exfiltration, remote script execution, or privilege escalation logic.
[SAFE]: The example code provided in the examples/ directory serves as an educational baseline for secure invoice processing and correctly demonstrates input validation and authorization checks.

typescript-engineering