The Agent Skills Directory

[PROMPT_INJECTION]: No malicious instructions, role-play attempts, or safety bypass patterns were detected. The instructions use natural language for professional guidance.
[DATA_EXFILTRATION]: There is no evidence of sensitive file access (such as credentials or environment variables) or network operations targeting external domains. Tools mentioned are standard local development utilities.
[REMOTE_CODE_EXECUTION]: The skill does not perform any remote script downloads, piped shell executions, or unsafe dynamic code generation. References to npm run build and lighthouse are routine for performance profiling.
[OBFUSCATION]: No Base64, zero-width characters, homoglyphs, or other encoding techniques were found in the skill content.
[INDIRECT_PROMPT_INJECTION]: While the skill involves processing performance measurement data (Lighthouse reports), it lacks dangerous capabilities that would make this an exploitation vector, and it serves an purely informational purpose.

performance-optimization