The Agent Skills Directory

[SAFE]: The skill outlines a standard performance optimization protocol (Target, Baseline, Profile, Hypothesis, Change, Re-measure, Record). This is a best-practice methodology for software engineering.
[EXTERNAL_DOWNLOADS]: The skill references several well-known and reputable performance measurement tools. These include Lighthouse (Google), hyperfine, and py-spy. All external links provided in the documentation lead to official GitHub repositories or established technical documentation sites (e.g., web.dev, postgresql.org).
[COMMAND_EXECUTION]: Instructions include the use of standard profiling commands such as py-spy record, node --cpu-prof, go test -bench, and hyperfine. These commands are used for their intended purpose of performance measurement and do not exhibit malicious patterns or unexpected privilege escalation.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to guide the agent's behavior when responding to performance-related queries. While it processes user-provided performance reports, it includes strict logic (the "Prime Directive") to prevent unjustified changes, which serves as a conceptual guardrail against following unverified instructions.
[SAFE]: References to other skills like godmode:fault-diagnosis and godmode:quality-gates indicate a modular design and do not present security risks.

performance-tuning