The Agent Skills Directory

[SAFE]: The skill serves as a documentation manifest for a tool in development and does not contain any executable scripts or malicious patterns.- [PROMPT_INJECTION]: A thorough analysis for instruction overrides, safety bypasses, and indirect injection markers was conducted; no such patterns were detected. While the skill's purpose involves processing untrusted external data (codebases and URLs), the manifest itself contains no malicious directives.- [DATA_EXFILTRATION]: The skill describes reading local files and accessing external URLs as part of its core reverse-engineering functionality. No unauthorized data harvesting or exfiltration techniques were identified.- [EXTERNAL_DOWNLOADS]: All external links provided in the documentation refer to legitimate research sources and design standards (e.g., Tokens Studio, W3C Design Tokens, C4 Model).

skill-extract