wai-yaml-ld: Accessibility Standards Data for Humans + LLMs

Machine-readable accessibility standards context (WCAG, ATAG, UAAG, ARIA, HTML, CSS) designed for humans and LLMs building standards-aligned code.

Experimental: This site now includes GraphQL-style static API support under /api (schema + introspection + JSON endpoints). LLM support varies by model and toolchain; some agents will use this data directly, others may not.

GraphQL-style API (Experimental)

Why this helps AI coding

Map code issues to exact Success Criteria (example: missing aria-label and SC 4.1.2).
Explain the “why” by tracing links from WCAG failures to related ATAG requirements.
Provide auditable evidence using linked standards data instead of generic best-practice claims.
Turn accessibility guidance into structured, queryable technical data.

How to test GraphQL support

Step 1: Open introspection.json to inspect available types and fields.
Step 2: Open a specific criterion endpoint (for example SC 4.1.2).
Step 3: Open all criteria and verify relationships (`parentGuideline`, techniques, failures).
Step 4: Use these URLs in MCP/AI tools as static GraphQL resolver targets over HTTPS GET.

Try this with your LLM

Prompt 1 (SC lookup): “Use /api/v1/introspection.json to discover fields, then fetch /api/v1/sc/4-1-2.json. Check my HTML for violations and explain how the issue maps to WCAG SC 4.1.2.”
Prompt 2 (traceable evidence): “Review this code and return findings with auditable evidence from wai-yaml-ld. For each issue, cite SC ID, linked failures/techniques, and explain the standards trace from WCAG to related ATAG context when available.”
Prompt 3 (experimental caveat): “If your runtime cannot directly use this API, say so explicitly and continue using the linked JSON artifacts as static references.”

Start Here

Interactive Exploration

Key Datasets

Governance & QA

If a link points to GitHub, it is intentional: these resources are source-of-truth repository artifacts.