narrative-builder
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes user-provided insights to generate stories without using boundary markers or sanitization. This creates an indirect prompt injection surface where malicious input could potentially influence the agent's use of the Write and Edit tools.
- Ingestion points: User-provided insights processed via storytelling triggers in SKILL.md.
- Boundary markers: Absent in the story templates provided in the skill instructions.
- Capability inventory: Read, Write, Edit, and AskUserQuestion tools.
- Sanitization: No evidence of input validation or escaping logic to handle potentially malicious user input.
Audit Metadata