# Axe Rules
> Load this skill when configuring axe-core scans, reviewing automated accessibility test results, or writing tests that use @axe-core/playwright, @axe-core/react, or similar integrations. Provides a quick reference to axe 4.x rule IDs, their WCAG mapping, and default severity levels.
- Skill: `mgifford/axe-rules` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add mgifford/axe-rules`
- Raw SKILL.md: https://api.skillmd.com/api/skills/mgifford/axe-rules/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Web & Frontend
- Author: mgifford (https://skillmd.com/u/mgifford)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/mgifford/axe-rules
---
# Axe-Core Rules Reference Skill
> **Canonical source**: `examples/AXE_RULES_REFERENCE.md` and `examples/AXE_RULES_COVERAGE.md` in `mgifford/ACCESSIBILITY.md`
> This skill is derived from those files. When in doubt, the example is authoritative.
> **Note:** the canonical files are organized around a specific test-fixture scanner
> project (numbered test pages). This skill re-presents the same rule set and
> severities as a general-purpose reference, independent of any particular fixture.
Apply these rules when writing, reviewing, or debugging axe-core scans.
---
## Core Principle
Axe-core catches ~30–40% of WCAG issues automatically. It is the baseline,
not the ceiling. Always pair automated scanning with manual keyboard and screen
reader testing.
**An axe-core result is evidence, not a conformance decision.** Preserve the
rule's raw outcome (`violations`, `incomplete`, `passes`, `inapplicable`)
exactly as axe reports it — do not overwrite or discard it when converting
to a project finding. Cite this file, `AXE_RULES_REFERENCE.md`, or the
rule's own `dequeuniversity.com/rules/axe/` page as the authority for a
WCAG mapping; do not invent a mapping axe-core does not document for that
rule. Record an unreviewed `violations` result as
`evidence_status: automated-indicator` with `handling: review` — never as a
confirmed standards failure — until a human confirms it. See
[Accessibility Finding Tracking: Policy Classification](https://mgifford.github.io/ACCESSIBILITY.md/examples/ACCESSIBILITY_FINDING_TRACKING.html#policy-classification).
**Key tags to include in scans:**
```typescript
.withTags(['wcag2a', 'wcag2aa', 'wcag21aa', 'wcag22aa'])
```
Axe-core 4.11 includes 100+ automated checks across these categories: WCAG 2.0
A/AA (68 rules), WCAG 2.1 A/AA (26 rules), WCAG 2.2 A/AA (1 rule, disabled by
default), Best Practices (30+ rules), WCAG AAA (3 rules, disabled by default),
Experimental (9 rules, disabled by default), and Deprecated (7 rules, disabled
by default).
---
## Severity Scale (axe-core)
| axe severity | Meaning |
|---|---|
| **critical** | Blocks access for one or more disability groups |
| **serious** | Significantly impairs access |
| **moderate** | Creates friction; workaround exists |
| **minor** | Best-practice gap; marginal impact |
---
## Critical Rules — must never appear in production
| Rule ID | Category | Description |
|---|---|---|
| `image-alt` | Text alternatives | `` missing `alt` attribute |
| `area-alt` | Text alternatives | Image map `` missing alt text |
| `input-image-alt` | Text alternatives | `` missing alt |
| `video-caption` | Text alternatives | `