frontend-coding-agent
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the 'mcporter' tool to execute Playwright automation commands for navigating to local or live URLs and capturing screenshots. This is part of a mandatory visual verification loop to ensure UI quality.
- [REMOTE_CODE_EXECUTION]: The 'sessions_spawn' function is used to delegate multi-file frontend builds to the 'Codex' sub-agent. This is a standard orchestration pattern for specialized coding agents.
- [DATA_EXFILTRATION]: The Playwright verification protocol references environment variables like '$NEXT_PUBLIC_APP_URL'. These are used to determine the target URL for automated browser testing and do not indicate unauthorized data transmission.
- [PROMPT_INJECTION]: The skill contains prescriptive design rules and operational protocols labeled as 'Non-Negotiable' and 'MANDATORY'. These instructions guide the agent's creative and technical output toward specific quality standards and do not attempt to bypass system safety constraints.
Audit Metadata