The Agent Skills Directory

[SAFE]: The skill provides instructional guidelines for an agent to evaluate music, lyrics, and meditations. It does not include any executable code, command-line operations, or remote dependencies. The stated purpose of scoring artifacts against a quality bar is consistent with the provided instructions.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external artifacts (such as lyrics and music briefs), which creates a potential surface for indirect prompt injection if those artifacts contain instructions intended to influence the agent. The risk is assessed as safe because the skill does not grant the agent dangerous capabilities like network access, and its output is limited to a markdown review file.
Ingestion points: Music artifacts, lyrics, meditation scripts, and Suno packets (SKILL.md).
Boundary markers: None defined to isolate external content from instructions.
Capability inventory: Capability to generate a review.md file and propose artifact revisions.
Sanitization: No specific input filtering or sanitization is mentioned.

music-taste-review