ui-visual-validator
Pass
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE] (SAFE): The skill consists entirely of markdown instructions and metadata. No executable scripts (.py, .js, .sh), configuration files, or binary assets are present within the provided scope.
- [NO_CODE] (SAFE): No code was found in the skill. It functions as a system prompt extension to define behavioral logic for UI testing and accessibility verification.
- [EXTERNAL_DOWNLOADS] (SAFE): The text mentions various professional UI testing tools (e.g., Chromatic, Percy, Applitools, BackstopJS) for context and knowledge-based guidance, but it does not contain commands to download, install, or execute these tools.
- [INDIRECT_PROMPT_INJECTION] (LOW): The skill is designed to process untrusted external data (UI screenshots and descriptions).
- Ingestion points: User-provided visual evidence and a referenced local file
resources/implementation-playbook.md. - Boundary markers: The instructions mandate a specific output format ('From the visual evidence, I observe...') which helps maintain structure.
- Capability inventory: None. The skill does not have file-write, network, or subprocess execution capabilities.
- Sanitization: Not applicable as the skill only produces natural language analysis.
Audit Metadata