The Agent Skills Directory

[SAFE]: The skill performs automated discovery by scanning standard project metadata files such as package.json, README.md, and CLAUDE.md. This information is used solely to generate a design brief and is consistent with the skill's stated purpose of creating context-specific UI specifications.
[SAFE]: Tool usage is restricted to standard file system operations (Read, Glob, Grep, Bash) for discovery purposes. No evidence of unauthorized network calls, sensitive credential harvesting (e.g., accessing .ssh or .env files), or data exfiltration was found.
[SAFE]: The instructions do not contain any obfuscation, prompt injection attempts, or persistence mechanisms. The skill generates markdown documentation and does not produce or execute hidden scripts.
[SAFE]: The skill mentions common industry libraries (e.g., axe-core, Playwright, D3, Recharts) as recommended tools for the resulting frontend implementation, but it does not perform any unverifiable package installations or remote script executions itself.

ui-brief