The Agent Skills Directory

[SAFE]: The skill is focused on educational quality and testing methodologies. It follows standard patterns for rubric development and conversation evaluation. The skill's primary function is to process pedagogical data and generate structured feedback.
[PROMPT_INJECTION]: The skill includes steps to process user-provided skill definitions and conversation transcripts. This ingestion of untrusted data represents a surface for indirect prompt injection. However, given that the skill has no specified high-privilege capabilities (such as network requests or file system modifications) and its output is limited to text-based evaluation, the associated risk is negligible. The agent is instructed to use this data strictly for analysis purposes.

skill-tester