ux-spec-author
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external design inputs such as wireframes, research documents, and mockups to generate structured specifications. This ingestion surface presents a theoretical risk of indirect prompt injection if source materials contain adversarial instructions intended to influence the agent's behavior. 1. Ingestion points: User-provided wireframes, research, and design system documentation. 2. Boundary markers: None; the instructions do not include specific delimiters or warnings to ignore instructions embedded within design inputs. 3. Capability inventory: The skill generates markdown documentation and interacts with other logical agent roles but does not perform shell command execution, network operations, or sensitive file access. 4. Sanitization: No explicit content filtering or validation mechanisms are defined for the input data.
Audit Metadata