ui-a11y
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes component or page source code (passed via the
$ARGUMENTSparameter) and is instructed to "Apply fixes directly where possible." This creates a vulnerability where malicious instructions embedded in the target source code (such as in comments, metadata, or string literals) could be interpreted by the agent, potentially leading to unauthorized code modifications or secondary injections. - Ingestion points: The skill ingests source code from components or pages specified by the user in the
$ARGUMENTSfield withinSKILL.md. - Boundary markers: There are no explicit delimiters or instructions defined to help the agent distinguish between legitimate code and potentially malicious instructions embedded within that code.
- Capability inventory: The skill provides the agent with the authority to perform file-write operations to apply "Auto-fixes" to the audited components.
- Sanitization: The instructions do not describe any sanitization, filtering, or validation steps for the input code before it is processed or modified.
Audit Metadata