ui-context-and-scope
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is entirely instructional and does not contain any executable scripts, shell commands, or network-enabled tools. All instructions provided are guidelines for human or AI UI design tasks.
- [NO_CODE]: No source code in Python, JavaScript, or other languages is provided within the skill, which eliminates risks associated with remote code execution, dynamic execution, or dependency vulnerabilities.
- [DATA_EXPOSURE_AND_EXFILTRATION]: No sensitive file paths, credentials, or environment variables are accessed. The skill includes positive security recommendations, such as advising against using CSS masking to hide sensitive data and suggesting pseudonymization before sending data to external models.
- [PROMPT_INJECTION]: The content does not contain instructions to override system prompts, bypass safety filters, or act in an unrestricted manner. The prompt signals and metadata are descriptive and aligned with the skill's stated purpose of providing UI context guidelines.
Audit Metadata