Skip to content
Back to skills

Run Typed Failure

ASecurity

실제 run으로 typed-failure 기록을 만들 때의 종결 상태 기대치

  • 17 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added September 5, 2026
ai-agentsgo

Works with

  • terminal

Security analysis

A100/100

Scanned September 5, 2026

npx -y skills add zzunkie/yardlet --skill run-typed-failure --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Run Typed Failure?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Run Typed Failure
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/zzunkie-run-typed-failure/badge)](https://www.skillsdirectory.com/skills/zzunkie-run-typed-failure)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: run-typed-failure
description: 실제 run으로 typed-failure 기록을 만들 때의 종결 상태 기대치
source: learned
---
실제 `yardlet run --task <id> --execute` 실패로 typed_run_failure_projection이 세는 run 기록(run.yaml state failed|partial + completed_at + evaluation.json fatal failed check)을 만들려면 goal feedback cap(max_feedback_cycles, 기본 2) 이내에서 실패를 반복하라. cap 이내 실패는 run을 state=partial로 봉인하고 태스크를 Partial(terminal)로 남긴다. cap을 초과하거나 max_feedback_cycles를 0으로 두면 feedback question이 붙어 태스크와 run이 needs_user로 파킹되어 projection에 잡히지 않는다. fixture 단언은 task state=failed가 아니라 partial을 기대해야 한다. 표준 예시: tests/fixtures/same_intent_replan/scripts/run.sh.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.