Prompt file imported from porsche-design-system/testing (.cursor/commands/a11y-quick-check.md). Fill in {{pageUrl}} before use. Copyright stays with the author.

Quick Web Accessibility Check

Name: a11y-quick-check
Author: porsche-design-system

Fast triage — run the deterministic Phase 1 scanner against a live URL and get a scoped verdict. No code review or screenshots.

Page to check

URL: {{pageUrl}}

Settings

Audit method: Runtime scan only (axe-core)
Thoroughness: Quick profile — phases 0 (pre-configured) → 1 → 11 only
Target standard: WCAG 2.2 AA
Report: Standard run artifacts plus an inline summary

Instructions

When the target provides the a11y-audit agent, use it. Otherwise execute this prompt directly with the installed a11y-* skills and preserve the workflow order and safety requirements below.

Use the quick profile:

Complete Phase 0 setup without discovery questions — settings are pre-configured above. Create $RUN (.a11y/runs/<YYYY-MM-DD-HHMMSS>) and $SCRATCH (mktemp -d), then read the finding schema and rule catalog before Phase 1.
Run axe-core against the URL as the first and only testing phase (Phase 1):
```
node <a11y-playwright>/scripts/a11y-scan.mjs \
  --url {{pageUrl}} --mode axe,tree,coverage \
  --out $SCRATCH/scan-axe-page-1.json
```
For an authenticated page, add --storage-state <file>. Do not use @axe-core/cli as the primary Phase 1 path. If scans.axe.status is not ok on a public page, run the CLI fallback without overwriting the primary scan:
```
npx --yes @axe-core/cli axe \
  {{pageUrl}} --tags wcag2a,wcag2aa,wcag21a,wcag21aa,wcag22aa \
  --save $SCRATCH/scan-axe-cli-page-1.json
```
Normalize both files.
Mark Phases 2–10 SKIPPED one at a time in numerical order because the quick profile omits domain and behavioral testing.
Normalize $SCRATCH/scan-axe-page-1.json and, when created, $SCRATCH/scan-axe-cli-page-1.json into $RUN/findings.json, run export-findings.py --format summary, and write $RUN/ACCESSIBILITY-AUDIT.md. Paste the generated numeric block unchanged as the inline Phase 11 summary. Add prose about scope, but do not restate any number or invent a second pass/fail formula.
Offer a full standard or deep-dive audit if the user wants remediation depth.
Retain the trimmed scan under $RUN/raw/, delete $SCRATCH, and print the report path.

Use the standard progress marker for every phase, including skipped phases, for example:

[A11Y AUDIT · Phase 1/12 · START] Automated baseline — running axe-core

# Quick Web Accessibility Check Fast triage — run the deterministic Phase 1 scanner against a live URL and get a scoped verdict. No code review or screenshots. ## Page to check **URL:** `{{pageUrl}}` ## Settings - **Audit method:** Runtime scan only (axe-core) - **Thoroughness:** Quick profile — phases **0 (pre-configured) → 1 → 11** only - **Target standard:** WCAG 2.2 AA - **Report:** Standard run artifacts plus an inline summary ## Instructions When the target provides the **a11y-audit** agent, use it. Otherwise execute this prompt directly with the installed `a11y-*` skills and preserve the workflow order and safety requirements below. Use the **quick** profile: 1. Complete Phase 0 setup without discovery questions — settings are pre-configured above. Create `$RUN` (`.a11y/runs/<YYYY-MM-DD-HHMMSS>`) and `$SCRATCH` (`mktemp -d`), then read the finding schema and rule catalog before Phase 1. 2. Run axe-core against the URL as the first and only testing phase (Phase 1): ```bash node <a11y-playwright>/scripts/a11y-scan.mjs \ --url {{pageUrl}} --mode axe,tree,coverage \ --out $SCRATCH/scan-axe-page-1.json ``` For an authenticated page, add `--storage-state <file>`. Do not use `@axe-core/cli` as the primary Phase 1 path. If `scans.axe.status` is not `ok` on a public page, run the CLI fallback without overwriting the primary scan: ```bash npx --yes @axe-core/cli axe \ {{pageUrl}} --tags wcag2a,wcag2aa,wcag21a,wcag21aa,wcag22aa \ --save $SCRATCH/scan-axe-cli-page-1.json ``` Normalize both files. 3. Mark Phases 2–10 `SKIPPED` one at a time in numerical order because the quick profile omits domain and behavioral testing. 4. Normalize `$SCRATCH/scan-axe-page-1.json` and, when created, `$SCRATCH/scan-axe-cli-page-1.json` into `$RUN/findings.json`, run `export-findings.py --format summary`, and write `$RUN/ACCESSIBILITY-AUDIT.md`. Paste the generated numeric block unchanged as the inline Phase 11 summary. Add prose about scope, but do not restate any number or invent a second pass/fail formula. 5. Offer a full standard or deep-dive audit if the user wants remediation depth. 6. Retain the trimmed scan under `$RUN/raw/`, delete `$SCRATCH`, and print the report path. Use the standard progress marker for every phase, including skipped phases, for example: `[A11Y AUDIT · Phase 1/12 · START] Automated baseline — running axe-core`

{ "prompt": "# Quick Web Accessibility Check\n\nFast triage — run the deterministic Phase 1 scanner against a live URL and get\na scoped verdict. No code review or screenshots.\n\n## Page to check\n\n**URL:** `{{pageUrl}}`\n\n## Settings\n\n- **Audit method:** Runtime scan only (axe-core)\n- **Thoroughness:** Quick profile — phases **0 (pre-configured) → 1 → 11** only\n- **Target standard:** WCAG 2.2 AA\n- **Report:** Standard run artifacts plus an inline summary\n\n## Instructions\n\nWhen the target provides the **a11y-audit** agent, use it. Otherwise execute this prompt directly with the installed `a11y-*` skills and preserve the workflow order and safety requirements below.\n\nUse the **quick** profile:\n\n1. Complete Phase 0 setup without discovery questions — settings are\n pre-configured above. Create `$RUN`\n (`.a11y/runs/<YYYY-MM-DD-HHMMSS>`) and `$SCRATCH` (`mktemp -d`), then read\n the finding schema and rule catalog before Phase 1.\n2. Run axe-core against the URL as the first and only testing phase (Phase 1):\n\n ```bash\n node <a11y-playwright>/scripts/a11y-scan.mjs \\\n --url {{pageUrl}} --mode axe,tree,coverage \\\n --out $SCRATCH/scan-axe-page-1.json\n ```\n\n For an authenticated page, add `--storage-state <file>`. Do not use `@axe-core/cli`\n as the primary Phase 1 path.\n If `scans.axe.status` is not `ok` on a public page, run the CLI fallback\n without overwriting the primary scan:\n\n ```bash\n npx --yes @axe-core/cli axe \\\n {{pageUrl}} --tags wcag2a,wcag2aa,wcag21a,wcag21aa,wcag22aa \\\n --save $SCRATCH/scan-axe-cli-page-1.json\n ```\n\n Normalize both files.\n\n3. Mark Phases 2–10 `SKIPPED` one at a time in numerical order because the quick profile omits domain and behavioral testing.\n4. Normalize `$SCRATCH/scan-axe-page-1.json` and, when created, `$SCRATCH/scan-axe-cli-page-1.json` into `$RUN/findings.json`, run\n `export-findings.py --format summary`, and write\n `$RUN/ACCESSIBILITY-AUDIT.md`. Paste the generated numeric block unchanged\n as the inline Phase 11 summary. Add prose about scope, but do not restate any\n number or invent a second pass/fail formula.\n\n5. Offer a full standard or deep-dive audit if the user wants remediation depth.\n6. Retain the trimmed scan under `$RUN/raw/`, delete `$SCRATCH`, and print the\n report path.\n\nUse the standard progress marker for every phase, including skipped phases, for example:\n\n`[A11Y AUDIT · Phase 1/12 · START] Automated baseline — running axe-core`", "variables": [ "pageUrl" ], "notes": "Quick accessibility triage of a web page. Runtime axe-core, structure, and coverage scan; no code review. Fast pass/fail verdict with score.", "metadata": { "source": { "provider": "github-cursor-commands", "repository": "https://github.com/porsche-design-system/testing", "path": ".cursor/commands/a11y-quick-check.md", "ref": "0e192ffa602d839ec70f42222b5c8b8e8b9daaf9", "url": "https://github.com/porsche-design-system/testing/blob/0e192ffa602d839ec70f42222b5c8b8e8b9daaf9/.cursor/commands/a11y-quick-check.md", "key": "porsche-design-system/testing/.cursor/commands/a11y-quick-check.md" } } }

a11y-quick-check

About

Quick Web Accessibility Check

Page to check

Settings

Instructions

Use it

Manifest

Reviews