AX Agent 도구·스킬 transcript 표현을 claw-code 기준으로 정리

- 도구/스킬 이벤트 배지를 역할 중심 라벨로 단순화
- raw snake_case 대신 사람이 읽기 쉬운 표시명과 /skill 형식 적용
- 작업 요약 팝업의 권한/배경 작업 관측성 섹션을 debug 수준으로 제한
- parity 문서와 개발 이력 문서에 tool/skill UX 정리 기준 반영

검증 결과
- dotnet build src/AxCopilot/AxCopilot.csproj -c Release -v minimal -p:OutputPath=bin\verify\ -p:IntermediateOutputPath=obj\verify\
- 경고 0 / 오류 0

docs/DEVELOPMENT.md

@@ -1,5 +1,8 @@
 # AX Copilot - 媛쒕컻 臾몄꽌
 
+- Document update: 2026-04-05 19:04 (KST) - Added transcript-facing tool/skill display normalization in `ChatWindow.xaml.cs`. Tool and skill events now use role-first badges (`도구`, `도구 결과`, `스킬`) with human-readable item labels such as `파일 읽기`, `빌드/실행`, `Git`, or `/skill-name`, instead of exposing raw snake_case names as the primary visual label.
+- Document update: 2026-04-05 19:04 (KST) - Reduced task-summary observability noise by limiting permission/background observability sections to debug-level sessions. Default AX Agent runtime UX now stays closer to `claw-code`, where transcript reading flow remains primary and diagnostics stay secondary.
+
 - Document update: 2026-04-05 18:58 (KST) - Switched `UserAskCallback` in `ChatWindow.xaml.cs` to a transcript-first inline card flow. `user_ask` no longer defaults to `UserAskDialog`; it renders an in-stream question card with choice pills, direct text input, and submit/cancel actions, then returns only the chosen answer to the engine.
 - Document update: 2026-04-05 18:58 (KST) - Aligned user-question UX with the same transcript-first principle already used for plan approval. `PlanViewerWindow` remains a secondary detail surface, while the primary approval/question decision now happens inside the AX Agent message timeline.
 

docs/claw-code-parity-plan.md

@@ -207,6 +207,16 @@
   - `plan approval`: transcript-first, detail view via `PlanViewerWindow`
   - `user ask`: transcript-first inline question card with choices / direct input / submit
 
+## Tool / Skill UX Parity Follow-up
+- Updated: 2026-04-05 19:04 (KST)
+- Default transcript should prefer role-oriented badges and readable labels over raw internal tool names.
+- AX implementation status:
+  - tool event badges: simplified to role-first labels
+  - item naming: normalized into readable Korean labels or `/skill-name` style
+  - observability panels: permission/background diagnostics reduced outside debug mode
+- Remaining quality target:
+  - move more tool-result and permission-result presentation into smaller message-type-specific helpers, closer to `claw-code` component separation
+
 ## Current Snapshot
 - Updated: 2026-04-05 19:42 (KST)
 - Estimated parity: