Uxaudit L3 Judge — Development agent for Claude Code
L3-vision Judge for the uxaudit pipeline.
How to install Uxaudit L3 Judge
Installs to ~/.claude/agents/gotalab-uxaudit-uxaudit-l3-judge.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/gotalab/uxaudit/HEAD/agents/uxaudit-l3-judge.md -o ~/.claude/agents/gotalab-uxaudit-uxaudit-l3-judge.md Restart Claude Code, or start a new session, for it to be picked up.
What Uxaudit L3 Judge does
name: uxaudit-l3-judge description: "L3-vision Judge for the uxaudit pipeline. Reads ONE captured screenshot plus a check-specific prompt.md (and optional evaluation brief) and writes a strict pass/fail/unverifiable verdict to result.json. Has Read and Write only — physically cannot drive a browser, run curl, or grep code. Invoked from Phase 04 Step 11 of the orchestrator playbook, one dispatch per L3 check, batched in parallel." tools: Read, Write model: opus
uxaudit L3-vision Judge
Alternatives in Development
- Vibe Coding Coach — Use this agent when users want to build applications through conversation, focusing on the vision and feel of 5.5k ★
- Eval Engineer — GAIA evaluation framework specialist 1.5k ★
- Build And Screenshot — 컴포넌트 메타데이터 검증/빌드, 프로젝트 빌드, 프로덕션 서버 시작, 스크린샷 캡처를 순차적으로 수행한다 395 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Uxaudit L4 Judge
L4-journey Judge for the uxaudit pipeline. Reads the per-journey capture directory (screenshots + steps.json +
OCR Reader
Batch OCR re-reader for low-confidence book pages. Reads page screenshot images via multimodal Read tool and e
Player Eval
Screenshot evaluation agent. Analyzes one screenshot and outputs structured JSON perception/decision.
Evaluator Teammate
Code evaluation teammate. Works in Agent Teams mode, receives Review Dispatch payloads, reviews scoped code ch
Skill Eval Reporter
Compares repeated paired execution results using blind A/B methodology and generates a skill effectiveness rep
Mechanical Reviewer
Fresh semantic milestone review over a frozen diff, producing a strict ledger-backed review result and correct
Related Skills
Paper Trail Audit Article
Per-citation audit of an academic article (LaTeX or Markdown). Produces RECEIPTS.md classifying each citation
Browserground
Local UI-grounding specialist for hybrid AI agents. Qwen3-VL-2B LoRA. Screenshot + text target → strict JSON b
Compile Screenshots
You are processing a vision-LLM screenshot batch report into the wiki.