The body is lean and terse, using compact bullets with no padded explanation of concepts Claude already knows; every line delivers a directive or named pattern.

The body gives specific, named guidance (e.g., "Protobuf, JSON minification", "gzip/br", "Set for lookups", "O(1) or O(n)") but no inline executable code examples — concrete code lives only in the referenced file, leaving the body's guidance directive rather than copy-paste ready.

The Workflow section sequences Baseline → Identify → Fix → Verify, and the Verify step ("Re-profile to confirm improvement and check for regressions") provides an explicit validation/feedback checkpoint.

The body is an organized overview that signals a single one-level-deep reference ("See [implementation examples](references/implementation.md)"), which is a real file, keeping navigation shallow and clear.

The description lists multiple concrete actions — "profiling bottlenecks, reducing latency, fixing memory leaks, improving throughput, or optimizing algorithm complexity" — matching the anchor for listing several specific actions rather than naming only a domain and some actions.

It states what ("Enforce universal standards for high-performance development") and gives an explicit when clause ("Use when profiling bottlenecks, reducing latency, fixing memory leaks, improving throughput, or optimizing algorithm complexity in any language"), satisfying both required elements.

Terms like "profiling bottlenecks", "reducing latency", "memory leaks", "throughput", and "optimizing algorithm complexity" are natural phrases a user would say when requesting performance work, giving good keyword coverage.

The triggers are reasonably distinct, but the framing "universal standards" and "in any language" is broad and could overlap with language-specific performance skills, so it does not clearly occupy an unambiguous niche.