The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external data which could theoretically contain instructions intended to influence the agent's behavior.
Ingestion points: Data is ingested from track lyrics, album metadata files, and a user-provided preference file named lyric-writing-guide.md (SKILL.md).
Boundary markers: The instructions do not define explicit delimiters to separate the data content from the agent's instructions, nor do they explicitly warn the agent to ignore instructions embedded in lyrics.
Capability inventory: The agent has the ability to Read, Edit, and Write files, enabling it to modify the project based on the data it processes (SKILL.md).
Sanitization: No specific sanitization or validation of the ingested text content is described before it is used to drive the refinement passes.

lyric-refiner