frontend-to-backend-requirements
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill generates documentation by interpolating user-provided requirements into a template.
- Ingestion points: User-provided feature descriptions, data needs, and action requirements found in SKILL.md.
- Boundary markers: None present; the skill does not use delimiters to separate user data from instructions.
- Capability inventory: The skill writes markdown files to the .claude/docs/ai/ directory.
- Sanitization: None present; the skill does not validate or sanitize the input content before writing to the file.
Audit Metadata