setup
/setup
Set up commit-vitals in the current project — choose the pages to measure, the engine, budgets and how commits are measured. Use when the user wants to start tracking Lighthouse / Core Web Vitals per commit, or asks to configure commit-vitals.
- Author
- odranoelBR ↗
- Repository
- commit-vitals ↗
- Licence
- MIT
- Last repository change
- 12.11.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 1
No signals detected. Repository changed on 12.11.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Set up commit-vitals in the current project — choose the pages to measure, the engine, budgets and how commits are measured. Use when the user wants to start tracking Lighthouse / Core Web Vitals per commit, or asks to configure commit-vitals.
Arguments
/setup [url ...]
Original author description
Set up commit-vitals in the current project — choose the pages to measure, the engine, budgets and how commits are measured. Use when the user wants to start tracking Lighthouse / Core Web Vitals per commit, or asks to configure commit-vitals.
How it works
- 1. Check the current state
- 2. Decide the pages
- 3. Pick the engine — explain the trade-off in one line each
- 4. Budgets and runs
- 5. Save
- 6. Measuring "the code of this commit"
- 7. Baseline
- 8. Wrap up
SKILL.md preview
---
name: setup
description: Set up commit-vitals in the current project — choose the pages to measure, the engine, budgets and how commits are measured. Use when the user wants to start tracking Lighthouse / Core Web Vitals per commit, or asks to configure commit-vitals.
argument-hint: "[url ...]"
---
# Set up commit-vitals
Goal: leave the project with a working `.vitals/config.json`, a first baseline
measurement, and the user knowing what happens on each commit.
User arguments (may be empty): $ARGUMENTS
## 1. Check the current state
Call the `vitals_status` MCP tool. If the project is already configured, show
the summary and ask what the user wants to change instead of starting over.
## 2. Decide the pages
- Use URLs from the arguments if given. Otherwise propose 2–4 pages that
represent the site's templates (home, a listing, a detail/product page, the
heaviest landing page), reading the router or sitemap if helpful. Ask the
user to confirm. One page per *template* beats many pages of the same kind.
- Give each a short `name` (`home`, `product`, `checkout`).
## 3. Pick the engine — explain the trade-off in one line each
| Engine | Use when | Note |
|---|---|---|
| `local` (default) | localhost, preview deploys, staging behind VPN | Runs Lighthouse on this machine via `npx`; needs Chrome. |
| `psi` | public production URLs | Google's PageSpeed Insights API; also returns real-user CrUX data. Set `PSI_API_KEY` for higher quotas. |
The PSI engine cannot reach localhost or private hosts; `vitals_init` will
reject that combination.
## 4. Budgets and runs
Defaults: score ≥ 90, LCP ≤ 2500 ms, CLS ≤ 0.1, TBT ≤ 200 ms, median of 3
runs. Keep the defaults unless the user has targets. Fewer than 3 runs makes
the history noisy.
## 5. Save
Call `vitals_init` with `urls`, `engine`, `form_factor`, `runs` and `budgets`.
## 6. Measuring "the code of this commit"
Explain briefly: right after `git commit`, a deployed URL still serves the old
code. Offer the option that fits their workflow:
- **Measure a local production build** (`local` engine + `http://localhost:…`
of `build && preview`, not the dev server — dev servers are 3–10× slower and
mean nothing).
- **Wait for the deploy**: if the site exposes the commit SHA in
`<meta name="commit-sha" content="…">` or a response header, edit
`.vitals/config.json` → `onCommit.waitFor`
(`{"type": "meta", "name": "commit-sha", "timeoutSec": 600}`). Offer to add
that meta tag to their layout if it is missing.
- **Don't measure on commit** (`onCommit.enabled: false`) and rely on the daily
schedule only.
## 7. Baseline
Call `vitals_measure` with `wait_seconds: 600` and show the result. This first
run is the baseline every later commit is compared to.
## 8. Wrap up
- Suggest committing the `.vitals/` folder (config + `runs/`) so the team
shares the history; `jobs/`, `logs/` and `report.html` are already
gitignored.
- Mention `/commit-vitals:schedule` for daily checks and
`/commit-vitals:report` for the chart.
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · odranoelBR/commit-vitals | 90 | MIT |
| HwangSlater/harness-fit | 100 | MIT |
| nesquikm/dev-process-toolkit | 100 | MIT |
| userKael/rastro | 100 | MIT |
| als-apg/osprey | 100 | BSD-3-Clause |
| dedujeong/gemba | 100 | Apache-2.0 |
| eigenwise/eigenwise-toolshed | 100 | MIT |
| awrshift/agent-memory-kit | 100 | MIT |
| lifecycle-innovations-limited/claude-ops | 100 | MIT |
Database snapshot: 13.11.2026 · package version f506b1786206 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 100regression-triage
Explain and fix a Core Web Vitals / Lighthouse regression reported by commit-vitals. Use whenever a commit-vitals result (vitals_result, vitals_measure, a hook message) shows a regression or a budget failure, or the user asks why performance got worse after a commit. - 100lighthouse-perf
Use when a web frontend needs Core Web Vitals (LCP, INP, CLS) checked on every PR, or its first performance budgets drafted. Sets up Lighthouse CI with Google's thresholds at the 75th percentile, per-route budgets by page type, resource-size budgets, report artifacts and PR comments. - 90audit-performance
Audit performance against Core Web Vitals and performance budgets when the user asks to check performance, audit speed, or analyze web vitals - 100core-web-vitals-check
Check Core Web Vitals (LCP, INP, CLS) from real Chrome visitors and the Lighthouse performance score for a list of pages or a whole sitemap, through Google PageSpeed Insights. Use when the user asks "check my Core Web Vitals", "PageSpeed test these pages", "why is my site slow… - 100react-performance
Runtime performance for React and Next.js; owns Core Web Vitals targets, bundle budgets and loading strategy. Use for measured speed problems - slow loads or interactions, Core Web Vitals or Lighthouse regressions, bundle size, images, fonts, caching and delivery, animation… - 100performance-budget
Use when reviewing CI configuration, bundle analysis workflows, or pull requests that add dependencies and assets. Distinguish bundle-size budgets from runtime budgets such as Lighthouse or Core Web Vitals so teams can enforce both.