Trust/autonomy 수치를 실제 기록으로 되짚는 검증법
Scanned 9/5/2026
Install to Claude Code
npx -y skills add zzunkie/yardlet --skill trust-autonomy --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Trust Autonomy?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/zzunkie-trust-autonomy)More formats (shields.io, HTML) on the badges page.
---
name: trust-autonomy
description: Trust/autonomy 수치를 실제 기록으로 되짚는 검증법
source: learned
---
trust 리포트를 검증할 때: (1) `yardlet trust --json`의 sources.runs_read를 `wc -l .agents/telemetry/runs.jsonl`과, transitions_read를 `.agents/transitions/*.yaml`의 records 총합과 대조해 '수기집계 아님'을 증명한다. (2) done_trust 4분류 합=task_instances, trustworthy_done_rate=evidence_backed/done_reached 항등식을 확인한다. (3) 개별 인스턴스를 telemetry 행(첫 시도가 Done인지)과 transitions의 Done→非Done 역행으로 스팟체크한다. (4) TUI/CLI 동일성은 두 표면이 같은 trust::report_text를 부르는지로 확인하면 렌더 diff 불필요.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!