跳到技能内容
词映 AI导航
发现工具选型指南Skills 技能AI 快讯AI图片AI视频AI文档AI办公AI助手AI搜索AI编程AI音频AI设计AI智能体模型与开发AI学习模型评测AI提示词内容检测AI金融AI 3D扫描与识别其他工具跨境电商 AI
词映 AI导航
工具库Skills 技能
我的收藏
词映 AI导航
发现工具选型指南Skills 技能AI 快讯AI图片AI视频AI文档AI办公AI助手AI搜索AI编程AI音频AI设计AI智能体模型与开发AI学习模型评测AI提示词内容检测AI金融AI 3D扫描与识别其他工具跨境电商 AI
Skills directory/KarryViber/Orb/claim-vs-reality-margin

Claude Code skill for checking whether a scheduled job actually ran

Add last-run evidence, a grace window, and a missed-run condition to a schedule so runtime behavior can be verified.

中文

Install in Claude Code

Provide Claude Code with a sample report schedule and its last successful run timestamp. Use a fake clock or offline records for the exercise.

  1. Review `profiles/.claude/skills/claim-vs-reality-margin/SKILL.md` and its requirements at the pinned GitHub commit.
  2. Create `.claude/skills/claim-vs-reality-margin/` in your project and save the reviewed file as `.claude/skills/claim-vs-reality-margin/SKILL.md`.
  3. Start Claude Code from that project's root, then select `/claim-vs-reality-margin` in the slash menu.

Example task

Use an hourly count report to exercise scheduled job verification. Supply its last success, next expected time, and a grace window. Check what happens when no success arrives.
`last_success = 08:00` · `expected_next = 09:00` · `grace = 5 minutes`

What to check

At `09:05` with no new success, report `missed run` and give the owner who should review it. A recorded success clears the alert.

Original SKILL.md on GitHub · Updated 2026-11-14

Frequently asked question

Why is a cron expression insufficient to verify the run?

The schedule expresses expectations and the run receipt shows what happened. Set a tolerable delay and compare the recorded timestamp to detect a miss.

Related skills from this repository

  • Review an external delivery draft in Claude Code
  • Claude Code skill for one source of truth across list settings
Browse Claude Code skills in the Chinese directory
词映 AI 导航Skills 技能
工具分类全部工具已本地化技能选型指南AI 快讯