blueprint
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from tickets and codebase facts to generate a blueprint, creating a potential multi-step injection chain.
- Ingestion points: The skill reads user-provided tickets, requirements, and explores the existing codebase in the Orient phase (SKILL.md).
- Boundary markers: There are no explicit delimiting or "ignore instructions" markers defined for the input data processing.
- Capability inventory: The skill writes markdown files to a local scratch directory (.scratch/) and invokes the
coding-standardsskill tool. It is explicitly instructed not to edit source or commit files. - Sanitization: There is no mention of sanitizing or escaping external content before it is interpolated into the blueprint template.
Audit Metadata