Skill

a11y-fix

Fix accessibility issues found by /a11y-audit. Use when asked to fix a11y issues, run /a11y-fix, or remediate accessibility findings.

npx claudepluginhub deventually/a11y-plugin --plugin a11y-wcag22

Tool Access

This skill uses the workspace's default tool permissions.

Preview

Fix accessibility issues identified by a prior `/a11y-audit` run. This is the user-facing entry point — it validates arguments and dispatches the agent.

SKILL.md

Similar Skills

applying-brand-guidelines

41.6k

Applies Acme Corporation brand guidelines including colors, fonts, layouts, and messaging to generated PowerPoint, Excel, and PDF documents.

3 files

anthropics-claude-cookbooks

creating-financial-models

41.6k

Builds DCF models with sensitivity analysis, Monte Carlo simulations, and scenario planning for investment valuation and risk assessment.

2 files

anthropics-claude-cookbooks

analyzing-financial-statements

41.6k

Calculates profitability (ROE, margins), liquidity (current ratio), leverage, efficiency, and valuation (P/E, EV/EBITDA) ratios from financial statements in CSV, JSON, text, or Excel for investment analysis.

2 files

anthropics-claude-cookbooks

Stats

Stars0

Forks0

Last CommitMar 31, 2026

Actions

View Source View Plugin View on GitHub View README

A11y Fix

Fix accessibility issues identified by a prior /a11y-audit run. This is the user-facing entry point — it validates arguments and dispatches the agent.

Usage

/a11y-fix — fix all automatable issues from the most recent audit

/a11y-fix check-id-001 — fix a specific check by ID

/a11y-fix --category "Forms and validation" — fix all fixable findings in a category

Dispatch

Always dispatch the agent. Never apply fixes yourself.

Step 1: Validate arguments

Check that at least one audit report exists in docs/a11y-reports/ using Glob for docs/a11y-reports/*-audit.json. If none found, print: "No audit report found. Run /a11y-audit first." and stop.

If specific check IDs are passed as arguments, load the latest audit JSON and verify each ID exists in the findings. If a check ID is not found, print: "Check ID '[id]' not found in the latest audit report." and stop.

If --category is passed, validate against the same 9 category names as /a11y-audit (case-insensitive, suggest closest on mismatch).

Step 2: Dispatch the agent

Invoke the a11y-agent agent using the Agent tool with subagent_type: "a11y-agent".

Pass the complete prompt:

Mode: fix Arguments: [the full original user arguments as-is] Check IDs: [specific IDs or "all"] Category: [parsed category or "all"]

Step 3: Stop

Do not proceed further. The agent handles everything: loading findings, prioritization, parallel fixing, verification, and reporting.

Rules

Never apply fixes. You are a dispatcher, not a fixer.

Always validate before dispatching. Catch bad arguments early.

Always dispatch the agent. Every invocation, no exceptions.

A11y Fix

Fix accessibility issues identified by a prior /a11y-audit run. This is the user-facing entry point — it validates arguments and dispatches the agent.

Usage

/a11y-fix — fix all automatable issues from the most recent audit
/a11y-fix check-id-001 — fix a specific check by ID
/a11y-fix --category "Forms and validation" — fix all fixable findings in a category

Dispatch

Always dispatch the agent. Never apply fixes yourself.

Step 1: Validate arguments

Check that at least one audit report exists in docs/a11y-reports/ using Glob for docs/a11y-reports/*-audit.json. If none found, print: "No audit report found. Run /a11y-audit first." and stop.
If specific check IDs are passed as arguments, load the latest audit JSON and verify each ID exists in the findings. If a check ID is not found, print: "Check ID '[id]' not found in the latest audit report." and stop.
If --category is passed, validate against the same 9 category names as /a11y-audit (case-insensitive, suggest closest on mismatch).

Step 2: Dispatch the agent

Invoke the a11y-agent agent using the Agent tool with subagent_type: "a11y-agent".

Pass the complete prompt:

Mode: fix
Arguments: [the full original user arguments as-is]
Check IDs: [specific IDs or "all"]
Category: [parsed category or "all"]

Step 3: Stop

Do not proceed further. The agent handles everything: loading findings, prioritization, parallel fixing, verification, and reporting.

Rules

Never apply fixes. You are a dispatcher, not a fixer.
Always validate before dispatching. Catch bad arguments early.
Always dispatch the agent. Every invocation, no exceptions.