Agent skill
a11y
(ePost) Use when "accessibility", "WCAG", "POUR", "a11y", "screen reader", or accessibility workflow ("audit", "fix", "review", "close finding") is mentioned across any platform — loads cross-platform a11y compliance rules
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/a11y-klara-copilot-epost-agent-kit
Metadata
Additional technical details for this skill
- keywords
-
a11y accessibility wcag aria screen-reader pour
- platforms
-
[ "all" ] - connections
-
{ "enhances": [] } - argument hint
- [--audit [platform] | --fix [#id|n] | --review [area] | --close [id]]
- agent affinity
-
[ "epost-a11y-specialist" ]
SKILL.md
A11y — Unified Accessibility Command & Foundation
Cross-platform accessibility foundation (WCAG 2.1 AA) with auto-detection of audit, fix, review, or close workflows.
Step 0 — Flag Override
If $ARGUMENTS starts with --audit: load /audit --a11y (references/a11y-workflow.md in audit skill). Pass remaining args as platform hint.
If $ARGUMENTS starts with --fix: load /fix --a11y (references/a11y-mode.md in fix skill). Pass remaining args as finding ID or count.
If $ARGUMENTS starts with --review: load /review --a11y (references/a11y.md in review skill). Pass remaining args as focus area.
If $ARGUMENTS starts with --close: load /audit --close (references/a11y-close.md in audit skill). Pass remaining args as finding ID.
Otherwise: continue to Auto-Detection.
Auto-Detection
Analyze $ARGUMENTS for positional patterns and keywords:
| Pattern | Dispatch |
|---|---|
audit or empty + staged changes exist |
/audit --a11y (detect platform from changed file extensions) |
fix or #<id> or bare number |
/fix --a11y with the ID/count |
review |
/review --a11y |
close or resolve |
/audit --close |
| Ambiguous | Ask user: audit, fix, review, or close? |
Platform Detection
| Signal | Platform | Skill |
|---|---|---|
.swift, .xib, SwiftUI |
iOS | ios-a11y |
.kt, Compose |
Android | android-a11y |
.tsx, .jsx, HTML, CSS |
Web | web-a11y |
If no signal, ask the user. Activate ONLY the skill matching the detected platform.
POUR Framework
| Principle | Requirement |
|---|---|
| Perceivable | Content available through sight, hearing, or touch |
| Operable | Interface navigable by keyboard, pointer, voice |
| Understandable | Content readable, predictable, input-assisted |
| Robust | Compatible with assistive technologies |
WCAG Conformance Levels
| Level | Requirement | Target |
|---|---|---|
| A | Minimum | Must pass |
| AA | Standard (legal baseline) | Our target |
| AAA | Enhanced | Nice to have |
Severity & Scoring
Accessibility score: 0–100 (start at 100, subtract per finding).
| Severity | Points | Examples |
|---|---|---|
| Critical | -10 | Missing labels, keyboard traps, no alt text, zero contrast |
| Serious | -5 | No focus indicator, non-descriptive links, missing headings |
| Moderate | -2 | Inconsistent nav, positive tabIndex, heading gaps |
| Minor | -1 | Missing ARIA on decorative elements |
PR Blocking Rule
Block PR when any of:
- 1+ critical violations
- 1+ regressions (resolved finding reappears)
- 5+ serious violations
Operating Modes
| Mode | Trigger | Behavior |
|---|---|---|
| Guidance | /review --a11y, direct questions |
Human-readable examples, no file writes |
| Audit | /audit --a11y |
JSON-only output, read-only (no Write/Edit) |
| Fix | /fix --a11y |
JSON status + code edits, surgical changes only |
| Close | /audit --close |
Update known-findings DB, JSON confirmation |
Routing
| Intent | Command | When |
|---|---|---|
| Audit violations | /audit --a11y |
"audit accessibility" |
| Fix violations | /fix --a11y |
"fix a11y", fix top N |
| Close finding | /audit --close <id> |
Mark resolved |
| Review compliance | /review --a11y |
"review accessibility" |
Data Store
.epost-data/a11y/
├── fixes/
│ ├── findings/ ← audit-a11y reports (audit-YYMMDD-HHMM.json)
│ ├── patches/ ← fix-a11y diffs (finding-{id}-YYMMDD.diff)
│ └── reviews/ ← review-a11y reports (review-YYMMDD-HHMM.json)
├── README.md
├── analysis.md ← trend summary
└── known-findings.json (v1.3 schema)
Schema: .claude/assets/known-findings-schema.json
Each finding: id, platform, wcag, title, file_pattern, code_pattern, fix_template, priority (1–3), resolved, resolved_date, fix_applied, source, first_detected_date.
On first run, any a11y command creates missing directories and copies README.md from assets template.
See data-store skill for directory convention and gitignore rules.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
majiayu000/claude-skill-registry
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
majiayu000/claude-skill-registry
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
majiayu000/claude-skill-registry
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
majiayu000/claude-skill-registry
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
majiayu000/claude-skill-registry
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
majiayu000/claude-skill-registry
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?