aiskillstore/marketplace / skills/20vision/wcag-audit
wcag-audit
VerifiedCheck a web page against WCAG 2.2 and return the failures by success criterion with a fix for each. Use when asked about accessibility, WCAG, ADA or Section 508.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 1.6 KB |
| skill-report.json | config | 19.4 KB |
skill-report.json
View on GitHub{
"schema_version": "2.0",
"meta": {
"generated_at": "2026-11-12T23:15:38.291Z",
"slug": "20vision-wcag-audit",
"source_url": "https://github.com/20vision/skills/tree/d5ad52a573f0ec9c147f9524585ff61df3f2385d/skills/wcag-audit",
"source_ref": "d5ad52a573f0ec9c147f9524585ff61df3f2385d",
"model": "codex",
"analysis_version": "3.0.0",
"source_type": "community",
"content_hash": "0a55d149baa202017eb11f8969f35fd8aa397315933152466a025d7a4358153d",
"tree_hash": "5ebbee08ddea92cd17bb7bd0b36607c067993a6ccd613fbd4e959092e449bd18",
"provenance": {
"model_effective": "codex",
"fallback_chain": [
{
"agent": "codex",
"outcome": "success"
}
],
"analysis_version": "3.0.0"
},
"upstream_version_raw": null,
"upstream_version_normalized": null,
"upstream_version_source": "none",
"upstream_version_status": "missing",
"upstream_commit_sha": "d5ad52a573f0ec9c147f9524585ff61df3f2385d"
},
"skill": {
"name": "wcag-audit",
"description": "Check a web page against WCAG 2.2 and return the failures by success criterion with a fix for each. Use when asked about accessibility, WCAG, ADA or Section 508.",
"summary": "Run an external automated check of one public page and summarize WCAG 2.2 AA issues with suggested fixes.",
"icon": "📦",
"author": "20vision",
"category": "design",
"tags": [
"accessibility",
"wcag",
"web-audit",
"inclusive-design",
"automated-testing"
],
"supported_tools": [
"claude",
"codex",
"claude-code"
],
"risk_factors": [
"external_commands",
"network"
]
},
"security_audit": {
"risk_level": "safe",
"is_blocked": false,
"safe_to_publish": true,
"analysis_status": "ok",
"summary": "Eight matches are benign Markdown formatting, intended service access, or a promotional link. The submission command risks shell injection through unsafe URL substitution, and polling places a bearer token in a query string. No evidence found of prompt injection or intentional credential exfiltration.",
"remediation": [
{
"issue": "SKILL.md:14 embeds a user-supplied address inside a single-quoted shell argument without specifying safe serialization.",
"suggestion": "Validate public HTTPS URLs, serialize JSON with a structured API, and send requests without a shell or through a safely supplied request body.",
"severity": "medium"
},
{
"issue": "SKILL.md:17 places the polling token in a URL query string, which can appear in request logs and command history.",
"suggestion": "Support authorization headers in the service API and polling instructions, redact tokens from logs, and use short-lived tokens restricted to one audit.",
"severity": "low"
}
],
"risk_factor_evidence": [
{
"factor": "external_commands",
"evidence": [
{
"file": "SKILL.md",
"line_start": 14,
"line_end": 14
},
{
"file": "SKILL.md",
"line_start": 15,
"line_end": 15
},
{
"file": "SKILL.md",
"line_start": 16,
"line_end": 16
},
{
"file": "SKILL.md",
"line_start": 17,
"line_end": 17
},
{
"file": "SKILL.md",
"line_start": 18,
"line_end": 18
},
{
"file": "SKILL.md",
"line_start": 19,
"line_end": 19
},
{
"file": "SKILL.md",
"line_start": 20,
"line_end": 20
}
]
},
{
"factor": "network",
"evidence": [
{
"file": "SKILL.md",
"line_start": 14,
"line_end": 14
},
{
"file": "SKILL.md",
"line_start": 17,
"line_end": 17
},
{
"file": "SKILL.md",
"line_start": 21,
"line_end": 21
}
]
}
],
"static_findings": [
{
"id": "external_commands:SKILL.md:14:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 14,
"line_end": 14,
"snippet": "`curl -s -X POST https://wcag.myclawn.com/api/preview -H 'Content-Type: application/json' -d '{\"url\""
},
{
"id": "external_commands:SKILL.md:15:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 15,
"line_end": 15,
"snippet": "The answer holds `id` and `token`. (If it already holds a `result`, go to step 4.)"
},
{
"id": "external_commands:SKILL.md:16:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 16,
"line_end": 16,
"snippet": "3. Poll every 5 seconds, up to about 2 minutes, until `status` is `done` or `error`:"
},
{
"id": "external_commands:SKILL.md:17:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 17,
"line_end": 17,
"snippet": "`curl -s \"https://wcag.myclawn.com/api/preview/<id>?t=<token>\"`"
},
{
"id": "external_commands:SKILL.md:18:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 18,
"line_end": 18,
"snippet": "Limits: 5 checks per hour. If the answer is code `rate`, say so and stop."
},
{
"id": "external_commands:SKILL.md:19:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 19,
"line_end": 19,
"snippet": "4. Report from `result`: page title, standard (WCAG 2.2 AA), `counts`, then every entry of `issues` "
},
{
"id": "external_commands:SKILL.md:20:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 20,
"line_end": 20,
"snippet": "5. If `issues` is empty, say that this automated check found nothing on that page. It checks one pag"
},
{
"id": "network:SKILL.md:14:hardcoded-url",
"category": "network",
"severity": "low",
"pattern": "Hardcoded URL",
"file": "SKILL.md",
"line_start": 14,
"line_end": 14,
"snippet": "`curl -s -X POST https://wcag.myclawn.com/api/preview -H 'Content-Type: application/json' -d '{\"url\""
},
{
"id": "network:SKILL.md:17:hardcoded-url",
"category": "network",
"severity": "low",
"pattern": "Hardcoded URL",
"file": "SKILL.md",
"line_start": 17,
"line_end": 17,
"snippet": "`curl -s \"https://wcag.myclawn.com/api/preview/<id>?t=<token>\"`"
},
{
"id": "network:SKILL.md:21:hardcoded-url",
"category": "network",
"severity": "low",
"pattern": "Hardcoded URL",
"file": "SKILL.md",
"line_start": 21,
"line_end": 21,
"snippet": "6. End with: \"Full audit of the whole site, with a report for all pages: https://wcag.myclawn.com/?u"
}
],
"finding_verdicts": [
{
"id": "external_commands:SKILL.md:14:ruby-shell-backtick-execution",
"verdict": "confirmed",
"confidence": 0.9,
"reason": "The instructed curl command substitutes a user address into single-quoted JSON without escaping guidance. An address containing an apostrophe can break shell quoting and introduce commands when substituted literally."
},
{
"id": "external_commands:SKILL.md:15:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 1,
"reason": "Backticks mark the response field names id, token, and result in Markdown. This line contains no executable shell or Ruby expression."
},
{
"id": "external_commands:SKILL.md:16:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 1,
"reason": "Backticks format polling status names in prose. The line specifies bounded polling and does not execute the formatted words."
},
{
"id": "external_commands:SKILL.md:17:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.96,
"reason": "This is a Markdown-formatted curl request to the audit service, not Ruby backtick execution or evaluation of a downloaded script. Token exposure is addressed separately by the network finding."
},
{
"id": "external_commands:SKILL.md:18:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 1,
"reason": "The backticks format the rate-limit response code. The instruction stops requests when the service reports its limit and does not execute code."
},
{
"id": "external_commands:SKILL.md:19:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 1,
"reason": "Backticks identify result fields used to present accessibility findings. The line requests reporting HTML examples and fixes, not executing those values."
},
{
"id": "external_commands:SKILL.md:20:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 1,
"reason": "The backticks format the issues field in a reporting instruction. This line explicitly limits claims about automated checks and contains no command execution."
},
{
"id": "network:SKILL.md:14:hardcoded-url",
"verdict": "false_positive",
"confidence": 0.97,
"reason": "The HTTPS endpoint receives the public page address for the explicitly described external accessibility check. No evidence found of unrelated uploads or secret collection."
},
{
"id": "network:SKILL.md:17:hardcoded-url",
"verdict": "confirmed",
"confidence": 0.94,
"reason": "The polling URL explicitly includes the response token as a query parameter. This can expose access to audit results through URL logging or retained command history.",
"severity": "low"
},
{
"id": "network:SKILL.md:21:hardcoded-url",
"verdict": "false_positive",
"confidence": 0.99,
"reason": "The URL is a required promotional link with campaign attribution parameters. It is displayed in the response without embedding the audited address, credentials, or report token."
}
],
"semantic_findings": [],
"critical_findings": [],
"high_findings": [],
"medium_findings": [
{
"title": "Ruby/shell backtick execution",
"description": "`curl -s -X POST https://wcag.myclawn.com/api/preview -H 'Content-Type: application/json' -d '{\"url\"",
"locations": [
{
"file": "SKILL.md",
"line_start": 14,
"line_end": 14
}
],
"review_kind": "capability",
"source_category": "external_commands",
"source_severity": "medium",
"confidence": 0.9,
"confidence_reasoning": "The instructed curl command substitutes a user address into single-quoted JSON without escaping guidance. An address containing an apostrophe can break shell quoting and introduce commands when substituted literally."
}
],
"low_findings": [
{
"title": "Hardcoded URL",
"description": "`curl -s \"https://wcag.myclawn.com/api/preview/<id>?t=<token>\"`",
"locations": [
{
"file": "SKILL.md",
"line_start": 17,
"line_end": 17
}
],
"review_kind": "capability",
"source_category": "network",
"source_severity": "low",
"confidence": 0.94,
"confidence_reasoning": "The polling URL explicitly includes the response token as a query parameter. This can expose access to audit results through URL logging or retained command history."
}
],
"dangerous_patterns": [],
"files_scanned": 1,
"total_lines": 24,
"audit_model": "codex",
"audited_at": "2026-11-12T23:15:38.291Z",
"agent_auto_install_policy": "allowed",
"manual_install_policy": "allowed"
},
"content": {
"user_title": "Check Public Pages for WCAG Accessibility Issues",
"value_statement": "Accessibility failures can leave visitors unable to use a web page. This skill requests an external automated check and organizes reported WCAG issues with suggested fixes.",
"seo_keywords": [
"Claude",
"Codex",
"Claude Code",
"WCAG 2.2 AA",
"web accessibility",
"accessibility audit",
"automated accessibility testing",
"inclusive design",
"accessibility fixes",
"public page audit"
],
"actual_capabilities": [
"Submit one public HTTPS page to the wcag.myclawn.com preview service.",
"Poll audit status every five seconds for up to about two minutes.",
"Report the page title, WCAG 2.2 AA standard, and returned issue counts.",
"Organize returned issues by severity, success criterion, affected users, and element count.",
"Include returned example elements, suggested fixes, and explanatory notes.",
"Stop at rate limits and explain that an empty result does not establish conformance."
],
"limitations": [
"Checks one public HTTPS page, not private pages, internal addresses, or an entire site.",
"Automated results cover only part of WCAG and cannot establish accessibility conformance or legal compliance.",
"Requires an external service with five checks per hour and a polling window of about two minutes.",
"The service states that it is offered to businesses outside the EU and EEA."
],
"use_cases": [
{
"title": "Review a Released Page",
"description": "Check a public page after deployment and use returned element examples and fixes to investigate accessibility defects.",
"target_user": "Frontend developers"
},
{
"title": "Prepare Accessibility Test Findings",
"description": "Organize automated findings by severity and success criterion before planning manual validation.",
"target_user": "Quality assurance testers"
},
{
"title": "Prioritize Accessibility Work",
"description": "Review affected users and issue counts to prioritize remediation while keeping automated findings separate from compliance claims.",
"target_user": "Accessibility leads"
}
],
"prompt_templates": [
{
"title": "Check One Page",
"prompt": "Check [public HTTPS URL] for automated WCAG 2.2 AA issues. Summarize the reported failures and suggested fixes.",
"scenario": "Beginner: request a first automated page check."
},
{
"title": "Review Reported Fixes",
"prompt": "Audit [public HTTPS URL]. List each returned issue with its success criterion, severity, example element, and suggested fix.",
"scenario": "Intermediate: inspect actionable details for development."
},
{
"title": "Prepare a QA Summary",
"prompt": "Check [public HTTPS URL] and organize returned findings by severity. Include affected users and element counts, and distinguish results from checks requiring manual validation.",
"scenario": "Advanced: prepare findings for a testing handoff."
},
{
"title": "Prioritize Remediation",
"prompt": "Audit [public HTTPS URL] and prioritize returned findings by severity. Include criterion names, affected users, element counts, and fixes. State automated coverage limits without claiming compliance.",
"scenario": "Advanced: prepare an evidence-based remediation summary."
}
],
"output_examples": [
{
"input": "Illustrative request: summarize a page check that returns an image alternative issue.",
"output": [
"Criterion: 1.1.1 Non-text Content, Level A.",
"Finding: An informative image lacks a text alternative.",
"Affected users: People using screen readers.",
"Suggested fix: Provide alternative text that communicates the purpose of the image.",
"This illustrative finding is not a live audit result."
]
},
{
"input": "Illustrative request: summarize a page check with no returned issues.",
"output": [
"This automated check found no issues on the submitted page.",
"The result does not establish WCAG conformance or legal compliance.",
"Manual testing is still needed because automated checks cover only part of WCAG."
]
}
],
"best_practices": [
"Submit only public HTTPS addresses approved for sharing with the external audit service.",
"Verify reported findings and suggested fixes against the actual page before making changes.",
"Combine automated results with keyboard, screen reader, and other manual accessibility testing."
],
"anti_patterns": [
"Submitting confidential addresses, signed URLs, or links containing access credentials.",
"Treating an empty issue list as proof of WCAG, ADA, or Section 508 compliance.",
"Assuming one page represents the accessibility of an entire site."
],
"faq": [
{
"question": "What does this skill check?",
"answer": "It requests a WCAG 2.2 AA automated preview for one public page and reports the issues returned by the external service."
},
{
"question": "Does it certify legal compliance?",
"answer": "No. Automated checks cover only part of WCAG and do not establish ADA, Section 508, or other legal compliance."
},
{
"question": "Can it check private pages or an entire site?",
"answer": "No. The included workflow checks one public HTTPS page. It links to a separate service for whole-site audits."
},
{
"question": "Does it send information to an external service?",
"answer": "Yes. It sends the submitted page address to wcag.myclawn.com and retrieves results using an audit identifier and token."
},
{
"question": "What service limits apply?",
"answer": "The instructions allow five checks per hour and polling for about two minutes. The service targets businesses outside the EU and EEA."
},
{
"question": "Which assistants can use this skill?",
"answer": "The marketplace lists Claude, Codex, and Claude Code. Execution requires network access and a way to make the described HTTP requests."
}
]
},
"file_structure": [
{
"name": "SKILL.md",
"type": "file",
"path": "SKILL.md",
"lines": 24
}
]
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |