The Agent Skills Directory

[SAFE]: The skill operates as a structured assessment tool using text-based prompts to conduct quizzes. It does not perform any sensitive file operations, network exfiltration, or unauthorized command execution.
[PROMPT_INJECTION]: The assessment prompts in quick-assessment-prompt.txt and comprehensive-assessment-prompt.txt contain clear instructional logic for the agent to follow. There are no attempts to bypass safety filters or override global system instructions.
[INDIRECT_PROMPT_INJECTION]: The skill ingests user input in the form of multiple-choice answers. While there are no explicit boundary markers, the narrow scope of the quiz and the lack of high-privilege tool usage (like network access or arbitrary file writing) render this interaction surface safe.
[OBFUSCATION]: A thorough review of all files, including the assessment prompts and README, found no evidence of Base64 encoding, zero-width characters, homoglyphs, or hidden URLs.

self-assessment