The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting external data and interpolating it into story documents.
Ingestion points: The idealPlan and newStoryDescription input parameters in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used during the data interpolation process.
Capability inventory: The skill utilizes the save_issue tool for Linear integration and the Write and mkdir commands for local file system operations.
Sanitization: Although the skill performs INVEST validation on the logic and structure of the stories, there is no technical sanitization of the input strings to prevent prompt injection or escape sequences.

ln-221-story-creator