design
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is entirely composed of Markdown-based design guidelines and best practices. It does not include any executable code, shell scripts, or binary files. The content is instructional and focused on UI/UX principles, accessibility, and framework-specific patterns.
- [NO_CODE]: No external packages or scripts are bundled with the skill. All logic is contained within natural language instructions for the AI agent, posing no direct execution risk to the host system.
- [METADATA_POISONING]: Metadata fields such as author and versioning are properly defined and consistent with the vendor identity (afonsoft). There are no attempts to hide malicious instructions or spoof other entities in the metadata.
- [INDIRECT_PROMPT_INJECTION]: While the skill defines a process for ingesting user briefs (
references/shape.md), it does not provide any automated tools or script-based processing that would create an exploitable attack surface. The risk remains at the standard low level inherent to processing user-provided text.
Audit Metadata