verify-ui
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses standard shell utilities like cat, grep, and lsof to inspect package.json files and identify active network ports for local development.
- [COMMAND_EXECUTION]: It manages local server processes by executing package manager commands (npm, yarn, pnpm, or bun) to start the application and the kill command to terminate them during cleanup.
- [COMMAND_EXECUTION]: The skill interacts with the agent-browser tool to perform automated UI actions, including navigation, screenshots, and accessibility checks.
- [PROMPT_INJECTION]: The skill processes untrusted data from web pages (via browser snapshots and console logs) and project configuration files. Ingestion points: browser-based tool output and local file reads. Boundary markers: none. Capability inventory: subprocess execution and browser control. Sanitization: none. This attack surface is inherent to the functional purpose of UI verification and visual testing.
- [SAFE]: No evidence of data exfiltration, hardcoded credentials, malicious obfuscation, or unauthorized remote code execution was found.
Audit Metadata