Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →community-access/markdown-a11y-assistant
Interactive markdown accessibility audit wizard. Runs a guided, step-by-step WCAG audit of markdown documentation. Covers descriptive links, alt text, heading hierarchy, tables, emoji (remove or translate to English), ASCII/Mermaid diagrams (replaced with full accessible text alternatives), em-dashes, and anchor link validation. Orchestrates markdown-scanner and markdown-fixer sub-agents in parallel. Produces a MARKDOWN-ACCESSIBILITY-AUDIT.md report with severity scores and remediation tracking.
Passed 18 of 22 test cases with the skill, and 4 of 22 without it.
and 16 more in the full report
14 fixed4 passed either way4 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
This benchmark measured an earlier version of the skill.
One of 14 test cases the skill fixed. Same model, same prompt, with and without the skill.
We previously ran an audit in this repository and created MARKDOWN-ACCESSIBILITY-AUDIT.md. We want to check our documentation again. Start the process.
| Checked | Without skill | With skill |
|---|---|---|
| Identifies the pre-existing MARKDOWN-ACCESSIBILITY-AUDIT.md report before asking new configuration questions. | Passed | Passed |
| Asks whether the user wants to start a fresh audit or continue from the previous audit report. | Missed | Passed |
The instructions your model reads when the skill loads.
You audit markdown documentation: link text, alt text, heading order, tables,
emoji, ASCII and Mermaid diagrams, em-dashes and anchor validity. You scan every
file in parallel, apply the fixes the user approves, and write a scored report.
HTML and JSX go to web-accessibility-wizard; Office and PDF go to
document-accessibility-wizard.
Get the scope, and get the two judgement calls the user owns:
Read .a11y-markdown-config.json if one exists and honour it over defaults.
Read the reference for a phase when you reach it, not before.
| Phase | What happens | Read |
|---|---|---|
| 0 | Scope, emoji mode, diagram mode, config discovery | references/phase-0-discovery.md |
| 1 to 2 | File discovery, then one scanner per file in parallel, then scoring | references/phases-1-3-scanning.md |
| 3 | Review gate: auto-fixes listed, judgement calls presented | references/phase-3-review-gate.md |
| 4 | Apply approved fixes, write the report, offer CSV export | references/phases-1-3-scanning.md |
Quality bar for the rewritten text: references/excellence-guidelines.md.
One file is one dispatch to markdown-scanner. Files are independent, so
dispatch them together. Use exactly this prompt shape. Never read a skill's
instructions into your own context, and never paste its body.
Activate the skill "markdown-scanner". If skill activation is unavailable,
read skills/markdown-scanner/SKILL.md and follow it.
Task: scan this file across all nine markdown accessibility domains.
Scope: <file path>
test cases passed, vs 4 / 22 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.