Menu
Use when multiple optimization candidates exist and you need to benchmark them consistently, compare impact and risk, pick the winner, and explain the outcome clearly enough for a demo or PR.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Use when implementing one focused performance hypothesis on a branch: make the smallest viable code change, measure it against the shared benchmark, and leave a clean handoff for evaluation.
Use when the team needs a repeatable benchmark or evaluation harness before parallel optimization work begins, so every candidate change is tested with the same command, data shape, and success criteria.
Use when you need to add lightweight performance instrumentation, capture a baseline, and identify the slowest endpoint or hot path with enough trace detail for planning and optimization.
Use when you have baseline trace evidence and need to turn it into ranked optimization hypotheses, parallel engineering workstreams, and a clear plan for which ideas should be tested first.
Use when the winning optimization is chosen and you need a crisp PR-ready snapshot: problem, evidence, benchmark table, code change summary, risk notes, and the exact story to tell in a demo.
Use when starting from an unfamiliar repository or GitHub URL and you need a fast performance-focused intake: how to run it, where the user-facing surfaces are, which endpoint is best to optimize, and what the next agents need.
| name | performance-evaluator |
| description | Use when multiple optimization candidates exist and you need to benchmark them consistently, compare impact and risk, pick the winner, and explain the outcome clearly enough for a demo or PR. |
Choose the winning branch with evidence.
This role is not just about the fastest number. Pick the best overall candidate for the hackathon: strong latency win, credible measurement, low regression risk, and a story that demos well.
