The Agent Skills Directory

[SAFE]: The skill instructions provide a comprehensive framework for a 'Design Director' persona, focusing on visual hierarchy, accessibility, and product-specific design. No prompt injection or behavior override patterns were found.
[COMMAND_EXECUTION]: The skill includes three Node.js scripts (scripts/a11y-static-check.mjs, scripts/design-audit.mjs, scripts/token-audit.mjs). These scripts use the standard node:fs module to read local project files and perform regex-based audits for accessibility and design tokens. These scripts are benign and do not execute external code or perform network operations.
[DATA_EXFILTRATION]: While the included scripts read local file content, they only output reports to the console. There are no network calls or mechanisms to transmit sensitive data externally.
[PROMPT_INJECTION]: The instructions contain several 'No-Ship Quality Gates' and 'Operating Principles' that restrict the agent's behavior to maintain quality, but they do not attempt to bypass platform safety filters or extract system prompts.
[REMOTE_CODE_EXECUTION]: There are no patterns involving the download and execution of remote scripts. The skill relies on built-in platform tools (like Browser or DevTools) as described in the documentation, which is within the expected execution environment.

frontend-design-director