audit-ui-states
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes
npxto download and execute@playwright/cli. This is a well-known testing utility provided by Microsoft, an established service provider. The usage is transparent and directly supports the skill's primary function of UI verification. - [COMMAND_EXECUTION]: The skill executes shell commands to run Playwright tests. These commands are static and intended to automate browser navigation for auditing purposes. The output is directed to a local evidence directory (
.playwright-mcp/), following standard testing practices. - [INDIRECT_PROMPT_INJECTION]: While the skill ingests external data by navigating to a target application (
$APP), the risk is minimal. The process is defined as a read-only audit that generates a fix plan and captures screenshots, rather than performing automated writes or high-privilege actions based on the ingested content. - [SAFE]: The instructions emphasize a read-only approach and require manual approval for any subsequent fix plans, ensuring human oversight and preventing unauthorized modifications to the codebase.
Audit Metadata