run-benchmark SKILL.md skillA
run-benchmark from onevcat/prowl is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add onevcat/prowl/run-benchmark@git:20261002.cc9cd4aPins git:20261002.cc9cd4a in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- onevcat/prowl
- Path
.claude/skills/run-benchmark/SKILL.md- Versions recorded
- 2
- Latest version
- git:20261002.cc9cd4a, committed Oct 2, 2026
- First version recorded
- Aug 2, 2026
- Size
- 9,739 bytes
- Repository stars
- 653
- License
- a license file GitHub does not identify
Run Prowl's performance benchmark layers for the current branch or a release, compare against recorded baselines, and write a persistent local report. Use only when the user explicitly invokes run-benchmark or explicitly asks to run performance benchmarks / verify performance impact. Do not invoke automatically after implementation work. Classify the user's intent first (hot-path change, new optimization, or release-level sweep), confirm it with the user, then run the matching layers.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Run Benchmark ## Invocation contract Run this skill only after an explicit user request. Implementing, fixing, or reviewing performance-adjacent code does not by itself authorize a benchmark run. All timing in this workflow assumes **one dedicated, quiet local machine** accumulating a series over time. Numbers from CI runners, other machines, or a loaded host are not comparable and must not enter the baseline series. Never run other builds, tests, or heavy agents concurrently with a timed run. Background and design decisions live in `docs-ai/057-performance-benchmark-suite/`; the measurement methodology (why workload context matters, why load kills comparability) in `docs-ai/032-performance-hardening/004-agent-detection-steady-state.md`. ## Step 0 — Preflight 1. `git status --short` — an A/B run switches the checkout to the baseline commit, so the tree must be clean and HEAD committed. If dirty, tell the user what is blocked and ask whether to commit/stash or fall back to a branch-only run. 2. Quiet-host check: compare `uptime` load1 against `sysctl -n hw.ncpu`. Above roughly half the cores, warn that starvation inflates numbers unevenly and ask whether to wait …
Read the whole file at its exact version.
How to install
mdr add onevcat/prowl/run-benchmark@sha256:e07a10246da215caThe line at the top of this page pins the label git:20261002.cc9cd4a, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_dx4zfjvddakduopl)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 2 versions of this file, between Aug 2, 2026 and Oct 2, 2026, about one every 61 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261002.cc9cd4a latest | 2026-10-02 | cc9cd4a | 9,739 B | A | view · diff |
| git:20260802.60b73ee | 2026-08-02 | 60b73ee | 9,728 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (9739 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
onevcat/prowl · 653 stars · license a license file GitHub does not identify · pushed 2026-10-20 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_dx4zfjvddakduopl GET https://markdownregistry.com/api/v1/resolve?ref=onevcat/prowl/run-benchmark GET https://markdownregistry.com/api/v1/blob/e07a10246da215cabca3ceed86358dc41b3d27f2a8dd2f482c3c7c2baf10baa4
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.