Menü
Compare baseline and candidate query performance metrics and declare a benchmark winner.
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Basierend auf der SOC-Berufsklassifikation
Designs dry-run experiments across query rewrites, indexes, workload twins, and rollout gates.
Produces a board-level briefing for closed-loop autonomous database operations.
Enforces closed-loop dry-run boundaries and blocks autonomous production apply.
Scores whether enough evidence exists to recommend, defer, or escalate an autonomous operator decision.
Compares do-nothing, dry-run-change, rollback, and defer scenarios for an operating objective.
Builds an executive-grade decision evidence packet from runtime signals and confidence checks.
| name | benchmark-ab-runner |
| description | Compare baseline and candidate query performance metrics and declare a benchmark winner. |
Run:
node runtime/runTool.js benchmark_ab_runner '{"baseline":{"p95Ms":300,"cpuMs":80},"candidate":{"p95Ms":190,"cpuMs":55}}'
Returns deltas, winner, and verdict.