AI agent skills for accessibility testing and WCAG 2.2 audits in Claude Code: planner, critic, Playwright/axe-core tester, screen reader reviews, bug reports, and draft ACRs.
This repository also configures its own agents. See what accessibility-skills tells them →
Agent Claude Code
Standalone accessibility design reviewer evaluating ARIA patterns, focus management, state communication, and multi-perspective access (screen reader, keyboard, low vision, cognitive). 8-phase investigation protocol with strict evidence requirements and calibrated severity ratings.
Agent Claude Code
Accessibility Design Planner — 9-phase protocol for designing accessible implementations before coding. Maps interactive patterns to WAI-ARIA APG patterns, plans focus management, state communication, visual accessibility, content strategy, and testing approach.
Agent Claude Code
ARRM-based accessibility role auditor — reviews artifacts from responsibility-based team role perspectives (visual design, UX design, front-end dev, content authoring, business analysis, testing) to produce role-attributed findings.
Agent Claude Code
Lightweight reconnaissance agent for accessibility reviews. Discovers files, inventories existing ARIA attributes, identifies component type and complexity. Returns structured summary for handoff to planner or critic.
Agent Claude Code
Updates benchmark documentation with results from bench-runner. Maintains BENCHMARK.md, README.md, and EVAL-GAPS-PLAN.md. Produces summary tables and analysis.
Agent Claude Code
Reviews eval suite quality — fixture/rubric consistency, scoring accuracy, false positive traps, difficulty calibration. Read-only reviewer that identifies issues without fixing them.
Agent Claude Code
Executes a11y skill benchmarks across hosted and local model families. Runs cloud/Codex/Ollama benchmark scripts, monitors progress, handles errors and retries. Reports raw results to the team.
Agent Claude Code
Creates and enriches eval suite fixtures — .md component files, .metadata.yaml grading criteria, and .rubric.yaml scoring definitions. Handles all three suites (critic, planner, perspective).
Agent Claude Code
Deep accessibility review from 7 access perspectives. Activated by escalation from a11y-planner or a11y-critic when one or more perspectives are flagged at MEDIUM or HIGH alarm level.
Agent Claude Code
Reads accessibility evidence artifacts (axe JSON, keyboard traces, SR censuses, source files, screenshots) and returns a contract-shaped digest plus an explicit coverage note. Extraction only — no severity, no recommendations, no fixes. Keeps evidence corpora out of the main context window. Spawn at haiku for pure…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: