By dequelabs
Deque's accessibility toolkit for coding agents. Get accessible UI effortlessly - your agent builds, tests, and remediates accessibility in the loop, so it's never a separate, manual chore. Currently powers the axe MCP Server analyze -> remediate -> verify workflow; expanding over time.
This skill should be used when the user asks to "audit accessibility", "fix all a11y issues on this page", "run the accessibility loop", "remediate accessibility until clean", "scan and fix localhost", or runs /axe-accessibility:mcp-audit. It drives the axe MCP analyze -> remediate -> apply -> verify loop on a URL, applying fixes each round until violations reach zero or a round cap is hit.
This skill should be used when the user asks to "generate accessibility instructions", "add axe instructions to my repo", "bake a11y into my coding agent", "create copilot-instructions for accessibility", "set up CLAUDE.md for axe", or runs /axe-accessibility:mcp-generate-instructions. It writes or updates agent-instruction files (CLAUDE.md, .github/copilot-instructions.md, Cursor rules, or AGENTS.md) that enforce the axe MCP analyze -> remediate -> verify workflow.
This skill should be used when the user asks to "set up axe MCP", "configure the axe MCP server", "install axe accessibility tooling", "connect axe MCP", "authenticate axe", or runs /axe-accessibility:mcp-setup. It interactively wires the axe MCP Server into the user's IDE/MCP client, helps choose and configure authentication (API key or OAuth), and verifies the connection.
This skill should be used whenever the axe MCP Server tools (analyze, remediate) are available and the task involves accessibility, a11y, WCAG conformance, or "fix accessibility issues" on web UI. It teaches the correct analyze -> remediate -> verify workflow, the exact field mapping between the two tools, and credit-aware usage. Trigger phrases include "check accessibility", "scan for a11y issues", "fix accessibility violations", "make this accessible", "axe scan".
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Deque's accessibility toolkit for coding agents — get set up fast and teach your coding agent to ship accessible UI.
v1 focuses on the axe MCP Server and its analyze and remediate tools, wrapped in a smooth onboarding + usage workflow. The plugin is the umbrella for Deque's agent-facing accessibility capabilities; more will be added over time.
| Component | Type | What it does |
|---|---|---|
mcp-usage | Skill (auto) | Background knowledge so any agent calls analyze/remediate correctly (field mapping, credit awareness, the workflow). Loads automatically on accessibility tasks. |
/axe-accessibility:mcp-setup | Skill (command) | Interactive setup: pick auth (API key or OAuth), configure your client, verify the connection. |
/axe-accessibility:mcp-generate-instructions | Skill (command) | Generate/merge agent-instruction files (CLAUDE.md, .github/copilot-instructions.md, Cursor rules, AGENTS.md) that enforce the workflow. |
/axe-accessibility:mcp-audit | Skill (command) | Drive the loop on a URL: analyze → remediate → apply → re-verify until 0 violations or a round cap. |
.mcp.json | MCP server | Ships an auth-agnostic axe MCP Server entry (Docker) that works with either API key or OAuth. |
dequesystems/axe-mcp-server:latest, pulled automatically on first launch (no docker login required).analyze/remediate tools will fail to authenticate.npx @deque/axe-auth).Run locally for development:
claude --plugin-dir /path/to/axe-accessibility
Install from the Deque-hosted marketplace (this repo doubles as one):
/plugin marketplace add dequelabs/axe-accessibility
/plugin install axe-accessibility
Once accepted into Anthropic's community marketplace, users can also:
/plugin marketplace add anthropics/claude-plugins-community
/plugin install axe-accessibility@claude-community
Two distribution paths, not mutually exclusive:
dequelabs; .claude-plugin/marketplace.json makes it installable immediately via /plugin marketplace add.claude plugin validate . (the review pipeline runs the same check plus automated safety screening), then submit at claude.ai/settings/plugins/submit or platform.claude.com/plugins/submit. Approved plugins are pinned by commit SHA in anthropics/claude-plugins-community and synced nightly. The curated claude-plugins-official marketplace is selected by Anthropic at its discretion (no application)./axe-accessibility:mcp-setup — choose API key or OAuth and connect the server. Verify with /mcp./axe-accessibility:mcp-generate-instructions all — bake the workflow into your repo's agent instructions./axe-accessibility:mcp-audit http://localhost:3000 — analyze, remediate, and verify until clean.Two mechanisms are supported; choose during /axe-accessibility:mcp-setup:
AXE_API_KEY.npx -y @deque/axe-auth login (browser PKCE flow; tokens stored in the OS keychain with auto-refresh).The bundled .mcp.json is auth-agnostic: it mints an OAuth token and passes exactly one credential — the OAuth AXE_ACCESS_TOKEN if you're logged in, otherwise AXE_API_KEY. (The server rejects having both set, so the config never passes both.) Set AXE_SERVER_URL for private cloud / on-prem deployments.
Note: Each
remediatecall consumes AI credits from your organization's allocation.analyzedoes not, so re-verifying is cheap.
MIT © Deque Systems, Inc.
npx claudepluginhub dequelabs/axe-accessibility --plugin axe-accessibilityReliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Browser automation and end-to-end testing MCP server by Microsoft. Enables Claude to interact with web pages, take screenshots, fill forms, click elements, and perform automated browser testing workflows.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.