The Agent Skills Directory

[SAFE]: The skill instructions and reference materials focus entirely on performance optimization and follow established best practices. No malicious patterns, obfuscation, or unauthorized access attempts were identified.
[EXTERNAL_DOWNLOADS]: The documentation recommends the use of standard development tools and libraries from the npm registry, such as webpack-bundle-analyzer, clinic, and autocannon. These references are contextually appropriate for performance auditing and do not point to untrusted or malicious sources.
[COMMAND_EXECUTION]: The skill involves executing shell commands for linting, testing, and performance profiling (e.g., autocannon -c 100, npx 0x app.js). These commands are standard for the domain and are used within a measurement-first workflow to provide verifiable performance improvements.
[PROMPT_INJECTION]: The skill includes clear operational boundaries, such as requiring baseline metrics before optimization and restricting changes to build configurations or architectural structures. These constraints act as a safeguard against unintended agent behavior.

bolt