ai-interface-design
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill establishes a robust process for AI interface design that prioritizes human-in-the-loop oversight, risk assessment (categorizing operations from R0 to R3), and transparency in AI capabilities and failures.
- [EXTERNAL_DOWNLOADS]: In the
principles.mdfile, the skill references authoritative resources from well-known and trusted organizations, including Microsoft HAX Guidelines, Google PAIR Guidebook, NIST AI RMF, and documentation from Anthropic and OpenAI. These references are used for informational purposes. - [COMMAND_EXECUTION]: The skill utilizes a local Python script
scripts/validate-review.pyto ensure that generated review documents follow the required format. The script performs static analysis (regex-based structural checks) on local markdown files and does not perform network activity or privileged operations. - [INDIRECT_PROMPT_INJECTION]: While the skill processes user-provided interface requirements, it mitigates risk by instructing the agent to never treat AI fluency as evidence of correctness and requires manual verification for high-risk (R2/R3) operations. The included validation script is limited to structural checks and does not execute content from the generated reports.
Audit Metadata