The Agent Skills Directory

[SAFE] (SAFE): Technical audit of all 48 files confirms that the skill is entirely composed of documentation, rule definitions (Markdown), and configuration metadata. There are no patterns suggesting prompt injection, data exfiltration, or malicious behavior.- [NO_CODE] (SAFE): No executable scripts, binaries, or automation code are included in the skill package. Build and validation instructions in the README are intended for development maintenance of the skill repository and do not execute in the agent's runtime environment.- [PROMPT_INJECTION] (SAFE): Instructions are strictly technical and oriented toward library usage. No bypass attempts, role-play injections, or instructions to ignore safety filters were detected.- [DATA_EXFILTRATION] (SAFE): No hardcoded credentials, sensitive file path access, or network exfiltration logic was found in the code examples or metadata.

nuqs