The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources such as build-notes.md and takes_packed.md to inform narration scripts. It lacks explicit boundary markers or sanitization logic to differentiate between legitimate data and potentially malicious instructions embedded within these files.\n- Ingestion points: Reads workspace/video-projects/<slug>/takes_packed.md and build-notes.md in SKILL.md.\n- Boundary markers: No delimiters or 'ignore instructions' warnings are present in the processing logic.\n- Capability inventory: The skill uses Read and Write tools for file system operations as defined in SKILL.md.\n- Sanitization: There is no evidence of input validation or instruction filtering for the ingested content.

script-writer