Benchmark optimization loop: turn "make it faster" into a bounded measured loop
Baseline first, one-hypothesis variants, benchmark each against a correctness gate, promote the fastest safe variant
- What
- Baseline first, one-hypothesis variants, benchmark each against a correctness gate, promote the fastest safe variant
- Cost
- Free
- Needs
- an operation to optimize and a correctness gate to protect; nothing else — the skill is a methodology, no accounts or keys required
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor — @affaan-m's discipline for turning "make it 20x faster" into a bounded, measured optimization loop instead of guesswork: no optimization until the baseline, the correctness gate, the metric (wall time, p95 latency, rows/sec, cost/run, memory, error rate) and the search budget exist; then measure the baseline, generate one-hypothesis variants, run each against the same input shape, reject anything that fails correctness, safety or reproducibility, promote the fastest safe variant, codify the winner, and re-run baseline vs. winner to confirm the delta. Includes a variant-tracking table, recursive/hyperparameter search rules (persist every run to a ledger, compare against the prior accepted winner, stop when improvement is within noise or cost exceeds budget), and a promotion gate — no variant becomes the default until correctness passes, the delta repeats, rollback is obvious, and the change is encoded in source control or a durable runbook. By @affaan-m, listed here with credit to its creator. Honest caveats: pairs with companion ECC lanes not included here; MIT-licensed. Skill Harbor never reviews the code, review it yourself before use. Discovered via skills.sh.
Version:
Install
Prerequisites: an operation to optimize and a correctness gate to protect; nothing else — the skill is a methodology, no accounts or keys required Install "Benchmark optimization loop: turn "make it faster" into a bounded measured loop" for me. Give my agent @affaan-m's benchmark-optimization-loop: define the baseline, correctness gate, metric and search budget before optimizing; measure the baseline, generate one-hypothesis variants, run each against the same input shape, reject variants that fail correctness/safety/reproducibility, promote the fastest safe variant, codify the winner, and re-run baseline vs. winner to confirm the delta — with the variant table and promotion gate. Repository: https://github.com/affaan-m/ecc/blob/main/skills/benchmark-optimization-loop/SKILL.md 1. Fetch the SKILL.md file (and any helper files) from the repository path into a temporary folder and summarize what it does in one or two sentences. 2. Safety check: review the SKILL.md and scripts for anything suspicious (unexpected network calls, shell commands, credential harvesting). This repo should contain zero secrets in code, credentials only via the secure vault, allowed hosts declared in the SKILL.md. Verify that holds here; STOP on any red flag and tell me. 3. Install it as a skill: copy SKILL.md and its helper files into the agent's skills directory, in a folder named "benchmark-optimization-loop". 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (e.g. nothing — it is a methodology; point it at a real operation with a correctness gate). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.