The Agent Skills Directory

[SAFE]: The skill is purely instructional and guides the agent in performing legitimate software testing tasks. No malicious code, obfuscation, or data exfiltration attempts were found.
[COMMAND_EXECUTION]: The skill utilizes standard, well-known development commands such as npm test and npx playwright test. These commands are used as intended for running automated test suites in a development environment.
[EXTERNAL_DOWNLOADS]: The skill references established industry services and documentation, including Playwright, Puppeteer, Chromatic, and Percy. These are well-known services in the frontend testing ecosystem and are documented neutrally as optional tools for visual verification.
[PROMPT_INJECTION]: No evidence of prompt injection, safety filter bypasses, or instruction-overriding patterns was detected. The strong instructional language used (e.g., "The Iron Law") serves to enforce best practices for UI testing rather than to subvert the agent's core instructions.

visual-regression-testing