wcag-diagnose
Skill4 starsChanged 5 days agoMITView on GitHub
Classify accessibility barriers against selected WCAG 2.2 or 2.1 criteria; route repairs by root cause, with WCAG 3.0 draft ideas strictly advisory. Use on unclear accessibility complaints, screenshots, keyboard and screen-reader defects.
In this repository55 other files it gives its agents.
- design-balance-and-proportionSkill
- design-brand-and-art-directionSkill
- design-buildSkill
- design-color-theorySkill
- design-diagnoseSkill
- design-hierarchy-and-emphasisSkill
- design-imagery-and-textureSkill
- design-layout-and-gridSkill
- design-movement-and-flowSkill
- design-reviewSkill
- design-rhythm-and-patternSkill
- design-typographySkill
- design-unity-and-systemsSkill
- design-validateSkill
- security-api-files-and-realtimeSkill
- security-authentication-and-sessionsSkill
- security-authorizationSkill
- security-browser-and-originSkill
- security-buildSkill
- security-configuration-and-secretsSkill
- security-crypto-and-dataSkill
- security-dependencies-and-buildSkill
- security-diagnoseSkill
- security-injection-and-validationSkill
- security-logging-and-failureSkill
- security-reviewSkill
- security-threat-model-and-designSkill
- security-validateSkill
- ux-accessibilitySkill
- ux-buildSkill
- ux-controls-and-inputsSkill
- ux-decisions-and-loadSkill
- ux-diagnoseSkill
- ux-feedback-and-performanceSkill
- ux-forms-and-errorsSkill
- ux-journeys-and-motivationSkill
- ux-navigation-and-structureSkill
- ux-onboarding-and-learningSkill
- ux-reviewSkill
- ux-trust-and-ethicsSkill
- ux-validateSkill
- ux-visual-hierarchySkill
- wcag-alternatives-and-mediaSkill
- wcag-assistive-tech-and-statusSkill
- wcag-buildSkill
- wcag-forms-and-error-preventionSkill
- wcag-keyboard-and-focusSkill
- wcag-language-and-predictabilitySkill
- wcag-navigation-and-discoverySkill
- wcag-pointer-and-touchSkill
- wcag-reviewSkill
- wcag-semantics-and-reading-orderSkill
- wcag-time-and-motionSkill
- wcag-validateSkill
- wcag-visual-and-responsiveSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 10
Read references/standard-selection.md and references/standards.json first.
- Very specific: line 19
When WCAG 3.0 is explicitly requested, read references/3.0/draft-guidance.json and report draft advisory observations separately, with the …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Diagnose an accessibility barrier (WCAG 2.1 / 2.2)
Scope
Read references/standard-selection.md and references/standards.json first. The W3C recommendation for the selected version is authoritative; catalogs are operational summaries. Use references/issue_taxonomy.json, followed by that version's criterion entries. Do not preload all the references or every specialist.
Procedure
- Establish task, affected page/route, component, state, input modality, device, viewport, browser and AT; distinguish observed evidence from inference. A screenshot cannot prove keyboard or screen-reader behavior.
- Resolve standard + level (explicit requirement > project policy > WCAG 2.2 AA). Cite that choice in the finding. Load
references/2.2/criteria.jsonfor 2.2 orreferences/2.1/criteria.jsonfor 2.1; exclude criteria above the level. The taxonomy spans versions and is not a conformance list by itself. - Classify the problem under one primary barrier category and relevant secondary categories. Map only to applicable criteria with evidence. States: confirmed failure, suspected failure needing manual verification, pass, not applicable, not tested.
- Differentiate closely related criteria using
references/cross_criterion_conflicts.json; consult version-matched Understanding and Techniques pages for exceptions. Techniques are informative, not required approaches. - Select relevant entries in
references/remediation_recipes.jsonand, if needed, the matching specialistreferences/playbook.md. Propose concrete repair and discriminatory verification, not an invented accessibility result. - When WCAG 3.0 is explicitly requested, read
references/3.0/draft-guidance.jsonand report draft advisory observations separately, with the dated snapshot; never blend them into the selected version's conformance matrix.
Result
Finding ID | selected WCAG version / level | observed symptom | route/state | evidence vs inference | applicable SC | classification | impact | practical repair | verification steps | confidence. Use references/finding.schema.json for JSON; the criterion must be valid for that version, not merely in the schema's combined enumeration.
4.1.1 Parsing is not a WCAG 2.2 criterion and is considered always satisfied for HTML/XML in the updated 2.1 recommendation. Fix harmful semantics under the relevant other criteria. Never assert legal compliance or automated certification.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.