critique-visual-hierarchy
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and critique external UI screens and rendered design mockups, which represent ingestion points for untrusted data. An attacker could embed malicious instructions within the design (e.g., as text labels) to manipulate the agent's critique output.
- Ingestion points: The skill ingests 'rendered screen' hierarchy and UI layout descriptions for analysis.
- Boundary markers: The skill does not define specific delimiters or instructions to ignore commands embedded within the design content.
- Capability inventory: There are no scripts, shell commands, network operations, or file system access points associated with this skill, ensuring that any potential injection cannot compromise the host system.
- Sanitization: No input validation or instruction-filtering mechanisms are present in the skill definition.
- [NO_CODE]: This skill consists exclusively of natural language instructions and does not include any executable code, scripts, or external package dependencies.
Audit Metadata