The Agent Skills Directory

[PROMPT_INJECTION]: No prompt injection or bypass patterns detected. The instructions are focused on guiding the user through a specific design methodology (the Measure area of the Decision Map).
[DATA_EXFILTRATION]: No sensitive file paths, credential access, or unauthorized network operations were identified. The skill references local documentation and external UX industry resources (e.g., Maze, Hotjar, User Interviews) for informational purposes.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns, script downloads, or dynamic execution mechanisms were found. The skill does not include any code files (.py, .js, .sh, etc.).
[COMMAND_EXECUTION]: No shell command execution or subprocess spawning detected. There are no instances of the dynamic context injection syntax (!command).
[CREDENTIALS_UNSAFE]: No hardcoded secrets, API keys, or private tokens were found in the metadata or instructional content.
[OBFUSCATION]: No obfuscated content, encoded strings (Base64, hex), or hidden characters were detected in any of the files.
[EXTERNAL_DOWNLOADS]: The skill provides links to reputable UX design and research websites for further reading. These are static links for the user's reference and are not used for automated downloads or execution.
[SAFE]: The skill serves as a structured workflow guide for UX researchers and designers, utilizing a canonical set of metrics and a specific questioning process without introducing technical security risks.

glare-measure-questioning