A local-first development skill to check keyboard accessibility, screen-reader semantics, contrast, focus states, and dynamic announcements on any UI surface.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install review-accessibility
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install review-accessibility using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Review Accessibility?
Review Accessibility is a highly targeted development skill engineered for AI coding agents to systematically evaluate and audit user interface surfaces for comprehensive digital accessibility. Operating entirely as a local-first and self-contained solution, it allows developers to quickly inspect UI elements, check keyboard operability, analyze ARIA roles, and ensure WCAG compliance without needing external connections or cloud environments.
Built to integrate seamlessly with Openclaw Skills workflows, this tool provides agents with strict guidelines to surface gaps, analyze contrast, and enforce logical focus orders. By keeping evaluations surgical and reversible, it empowers development teams to maintain high accessibility standards directly within their localized workspace and repositories.
Review Accessibility Use Cases
- Automated UI Accessibility Audits: Systematically scanning a freshly built component to verify keyboard-only paths and screen-reader compatibility.
- WCAG AA Compliance Testing: Inspecting text and crucial UI elements to ensure they meet essential color contrast ratios.
- Focus State and Semantic Verification: Auditing visible focus indicators, logical tab sequences, and landmark roles across dynamic web pages.
- Pre-flight Commit Checks: Running local assertions via Openclaw Skills to confirm that dynamic UI updates properly announce changes via aria-live or specialized roles before merging code.
How Review Accessibility Works
- Context Analysis: The AI agent ingests the UI component source code or content specified in the prompt.
- Requirement Checklist Parsing: The skill systematically evaluates five core pillars: keyboard reachability, ARIA/semantic roles, focus order/visibility, text contrast, and dynamic announcements.
- Gap Identification: The agent identifies exact offending elements for any item that fails to meet compliance guidelines.
- Surgical Remediation: Following a strict 'think before you edit' principle, the skill isolates the smallest necessary code changes to resolve gaps without introducing scope creep.
- Verification & Logging: The agent clearly separates confirmed fixes from assumptions, generating a comprehensive report or structured log of the accessibility posture.
Review Accessibility Setup
This skill functions completely out-of-the-box in standalone mode with zero configuration required. To expand its capabilities, use Openclaw Skills cloud upgrades for deeper workspace discovery and team alignment.
To optionally initialize cloud workspace capabilities, install the interface utility:
npm install -g @ritualai/cli
ritual init
ritual status
Review Accessibility Data Schema & Taxonomy
The skill generates structure and documentation inside your repository to preserve accessibility insights over time.
Open Knowledge Format (OKF) Schema
When reusable engineering decisions or accessibility patterns are discovered, the skill can generate a portable Markdown file with YAML frontmatter saved to knowledge/engineering/<slug>.md:
| Field | Type | Description |
|---|---|---|
type |
String | The type of convention (e.g., API Convention, Accessibility Standard). |
title |
String | A brief, one-line title of the reusable knowledge. |
description |
String | A concise line stating what was learned and during which task. |
resource |
String | Relative file path or pointer to the evidence inside the codebase. |
tags |
Array | Relevant taxonomy keywords for categorization. |
timestamp |
String | ISO 8601 formatted string of when the knowledge was captured. |
Review Accessibility Advanced Features
- Local-First Standalone Execution: Operates completely in your local environment, making it secure and fully functional without an internet connection or external SDKs.
- Optional Knowledge Capture: Automatically offers to write versionable Open Knowledge Format (OKF) files into the repository to make the codebase inherently smarter for future agent interactions.
- Ritual Cloud Upgrades: Easily connects to cloud environments for deep workspace discovery, multi-component prioritization, visual/regression test analysis, and team-ready remediation summaries.
- Strict Working Principles Enforcement: Guarantees non-destructive edits by enforcing surgical, reversible changes and explicit confirmation tracking.
SKILL.md
Loading
Related Openclaw Skills
Review Error Handling
A local-first development skill that audits codebases for robust error paths, input boundaries, timeout configurations, structured logging, and operation idempotency.
GoCreative Compliance
An on-demand, pay-per-call compliance toolkit enabling AI agents to screen entities against global watchlists, sanctions, and PEP databases instantly.
Review Loading, Error, Empty States
A local-first development skill to audit and ensure proper loading, error, empty, partial, and retry states across UI surfaces.
GoCreative Enrichment
A pay-per-call B2B enrichment skill for AI agents that extracts firmographics, work emails, domain intelligence, and social profile data using x402 auto-payment.
React Component Review
A local-first development skill designed to review React and TypeScript components for correctness, hook safety, prop modeling, and accessibility with minimal, precise changes.
Performance Regression Triage
A structured workflow for AI agents to systematically isolate and triage system performance issues using concrete telemetry data.
Featured*
