jubscodes/get-shit-pretty/gsp-accessibility

Claude Code and agent skill by jubscodes/get-shit-pretty

Trust tier 2
  • community
  • ext:[{"type":"hook","description":"Pre/post lifecycle hooks"}]
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Audit accessibility — contrast, tokens, screens, code, or compliance statement. Picks the right mode from project state when invoked with no flags — use when: is this accessible, check contrast, WCAG check, a11y audit, accessibility report, before we ship, accessibility statement

Install

npx vskill@latest install jubscodes/get-shit-pretty/gsp-accessibility

Semi-extensibleMentions customization, but needs skill-specific knowledge. Read the extensible skills standard

SKILL.md

<context> Unified accessibility skill — every WCAG mode lives here. Merges the former gsp-accessibility (quick checks, token audit, pre-emit validate gate) and gsp-accessibility-audit (design audit, code audit, compliance statement) into one context-aware router.

Two consumption patterns:

  1. Standalone — user runs /gsp-accessibility and the skill picks the right mode from project state, or runs an explicit --mode
  2. As a building block — gsp-brand-guidelines calls --validate as a pre-emit gate; gsp-project-critique and gsp-project-review reuse the chunks this skill writes

Accessibility runs at multiple checkpoints in the pipeline (brand emit, post-design, post-build, pre-ship). The mode determines which checkpoint. </context>

<objective> Run the right accessibility check for the project's current state.

Input: Mode flag (or none — auto-pick) + optional arguments Output: Display, chunk file, statement, or audit log depending on mode Agent: gsp-accessibility-auditor (only in --design and --code modes) </objective>

<rules>

  • Always use AskUserQuestion for user interaction — never prompt via plain text
  • One decision per question — never batch multiple questions in a single message
  • Quick check mode (--check) and validate mode (--validate) never write chunk files
  • --validate carries verdict via exit code; do not swallow exit 1
  • Default conformance level is AA unless overridden by --level AAA or project config
  • Foundation chunks follow chunk-format.md format

</rules>

<process>

Step 1: Parse invocation

Read $ARGUMENTS to determine the mode:

InputModeSiblingAgent?Output
--check #FG #BGQuick contrastmodes/check.mdNoDisplay only
--validate <yml>Pre-emit WCAG gatemodes/validate.mdNoStdout verdict + <yml-dir>/wcag-validate.log
--tokensBrand token auditmodes/tokens.mdNocritique/accessibility-token-audit.md
--designDesign screen auditmodes/design.mdYescritique/accessibility-audit.md + fixes
--codeCodebase auditmodes/code.mdYesreview/accessibility-audit.md + fixes
--statementCompliance statementmodes/statement.mdNoexports/accessibility-statement.md
(no args)Context-aware default—dependsdepends

Additional flag: --level AAA overrides conformance level (default: AA).

Step 2: Context-aware default (no args)

When invoked with no mode flag, pick the right mode from project state:

  1. No `.design/projects/` and no yml path → mode picker via AskUserQuestion:
  • "Quick contrast check (--check)"
  • "Audit a brand token file (--tokens)"
  • "Validate a brand yml (--validate)"
  • "Run a full audit on this project (resolves below)"
  1. Brand mid-emit (`gsp-brand-guidelines` is the caller) → --validate {yml} (callers pass it explicitly; this clause is a safety net)
  2. Project exists with design phase complete and no `critique/accessibility-audit.md` → --design
  3. Project exists with build phase complete and no `review/accessibility-audit.md` → --code
  4. Project exists with both audits present → offer --statement
  5. Project exists with neither phase complete → tell user to complete a phase first; offer --tokens if a brand exists

Once the mode is resolved (explicit or auto-picked), continue to Step 3.

Step 3: Route to mode

Load the matching sibling and follow its instructions:

  • ${CLAUDE_SKILL_DIR}/modes/check.md
  • ${CLAUDE_SKILL_DIR}/modes/validate.md
  • ${CLAUDE_SKILL_DIR}/modes/tokens.md
  • ${CLAUDE_SKILL_DIR}/modes/design.md
  • ${CLAUDE_SKILL_DIR}/modes/code.md
  • ${CLAUDE_SKILL_DIR}/modes/statement.md

Each mode file is self-contained: it resolves its own context (project, brand, conformance level), describes its checks, and specifies its output. Agent-spawning modes (--design, --code) also reference ${CLAUDE_SKILL_DIR}/wcag-checklist.md and ${CLAUDE_SKILL_DIR}/methodology/gsp-accessibility-auditor.md.

Step 4: Update STATE.md (chunk-writing modes only)

After --tokens, --design, or --code finish, if running within a project and chunk files were written:

  • Read {PROJECT_PATH}/STATE.md
  • Note accessibility audit completion in the relevant phase section
  • Do not change phase status — accessibility is a supplementary check across phases, not a phase itself

--check, --validate, and --statement do not update STATE.md (no chunk written, or output is final artifact not phase progress). </process>

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
43
Repository forks
4
Repository npm downloads, 7 days
59
7-day trend score
32

Stars and forks are for the whole GitHub repository, not this skill alone. Downloads are for the repository's npm package. Refreshed 17 May 2026, 5 months ago.

Extension points

  • hookPre/post lifecycle hooks

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/jubscodes/get-shit-pretty/gsp-accessibility/badge)](https://verified-skill.com/skills/jubscodes/get-shit-pretty/gsp-accessibility)

Something wrong with this skill or its page? Tell us.