story-idea-generator
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied story ideas, genres, and preferences to generate refined concepts, character sketches, and pitches. These are then saved to the local file system using user-influenced filenames.
- Ingestion points: User input during the generation phases (e.g., "I want to write sci-fi") and concept names provided by the user.
- Boundary markers: The instructions do not define explicit delimiters or instructions to the agent to treat the processed story ideas strictly as data, increasing the risk of the agent following instructions embedded within the user's story text.
- Capability inventory: The skill utilizes file-writing capabilities to persist output in
context/output-config.md,.story-idea-generator-output.md, and dynamically named files such as{concept-name}-{date}.md. - Sanitization: There is no evidence of sanitization or validation for user-provided concept names used in file paths, nor for the content written to the files.
Audit Metadata