experience-architect
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, including product source code and rendered UI content, to diagnose UX problems and perform repairs. This establishes a surface for indirect prompt injection where instructions embedded in the analyzed code or UI could attempt to influence the agent's behavior.
- Ingestion points: The skill reads source code, markup, and styles (
references/_shared/evidence-standard.mdLevel E3) and observes rendered output in browsers or terminals (Level E1). - Boundary markers: The instructions do not explicitly mandate the use of delimiters or 'ignore instructions' warnings when the agent interpolates external code into its context, although it does include rules to prioritize product goals over user sentiment.
- Capability inventory: The skill possesses file-writing capabilities in 'REPAIR' and 'BUILD' modes and can trigger rendering actions in a browser or terminal environment.
- Sanitization: There is no specific requirement for sanitizing or escaping external content before it is processed by the agent.
Audit Metadata