The Agent Skills Directory

[SAFE]: The skill facilitates a legitimate user-driven workflow for content derivative generation. File system interactions (Glob search, Read, and Write) are limited to production-specific directories like '01-内容生产' and '02-已发布内容', posing no threat to system integrity.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests data from external markdown files and passes this content to the 'li-writer' skill. Ingestion points: Script files retrieved from '01-内容生产/待发布的内容/中短视频/' and '02-已发布内容/短视频/'. Boundary markers: None identified in the prompt templates. Capability inventory: File system access and skill orchestration. Sanitization: None, as the skill assumes source scripts are trusted content for transformation.

li-distribute