sk Skill · affaan-m
benchmark-optimization-loop
Use when the user asks to make something faster, try many variants, run recursive optimization, benchmark latency/throughput/cost, or choose the best implementation by repeated measured tests.
Open on skills.sh ↗read 2026-09-17
- installs 8w
- 1,726
- 30-day movement
- starts with the next reading
- Related entries
- 2
- Connections
- 6
JavaScript
A SKILL.md that turns vague speed-up requests into a bounded, measured optimization loop: baseline first, bottleneck analysis, one-hypothesis variants, and a variant table. It defines a promotion gate so a variant only becomes the new default after passing measured tests.
Reach for it when a user asks to make something faster or run many variants, so the agent optimizes by evidence instead of guessing.
Use it to
- Benchmark latency, throughput, or cost
- Run recursive or hyperparameter optimization
- Compare implementation variants by measured tests
- Establish a required baseline before optimizing
- Track variants in a table with promotion gates
For Agent users asking for performance or optimization work
- Host repository
- affaan-m/ECC
- Installs, lifetime
- 2,900
- Installs, 8 weeks
- 1,726
- Allowed tools
- Read, Write, Edit, Bash, Grep, Glob
- Licence
- MIT
- Host stars
- 261k
- Host language
- JavaScript
topicsbenchmarkingoptimizationagentsskillsperformancemeasurement