The Agent Skills Directory

[SAFE]: The skill demonstrates secure design by requiring explicit user approval for generated content before writing to the filesystem. It also incorporates validation logic to ensure only appropriate data sources (streaming lyrics vs. AI phonetics) are used for public-facing copy.\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its core function of processing external album content into generative prompts.\n
Ingestion points: The skill ingests album narratives, track concepts, and lyrics via the get_album_full method of the bitwize-music-mcp tool.\n
Boundary markers: No specific delimiters or boundary markers are defined for the input data in the generation phase.\n
Capability inventory: The skill uses Read, Edit, and Write tools to manage the promo/ directory.\n
Sanitization: There is no automated sanitization of the ingested lyrics; however, the risk is mitigated by a mandatory manual review step where the agent must present all generated copy for user approval before writing files.

promo-writer