The Agent Skills Directory

[Prompt Injection] (SAFE): No bypass markers, role-play instructions, or attempts to override system safety guidelines were detected.
[Data Exposure & Exfiltration] (SAFE): No hardcoded credentials, sensitive file paths (e.g., ~/.aws), or unauthorized network communication patterns were found.
[Obfuscation] (SAFE): The documentation and logic are written in clear text with no Base64, zero-width characters, or homoglyph-based evasion.
[Unverifiable Dependencies & Remote Code Execution] (SAFE): The skill does not reference or install external packages (npm, pip) and contains no logic for downloading or executing remote code.
[Indirect Prompt Injection] (SAFE): While the skill describes ingesting URL parameters for state management, it lacks the exploitable sinks (like shell execution or file writing) required for an attack.

variant-selection