verification-before-completion — 完成前证据核验

SkillFiles & storage

Use before EZ_math_model declares quality_audit, packaging, or final delivery complete, especially when pass/fail claims must be backed by files, commands, counts, or concrete evidence.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the verification-before-completion — 完成前证据核验 skill

What this skill tells your AI

The instructions your AI receives, as published by woodfishhhh/ez_math_model in skills/ez-math-model/tools/verification-before-completion/SKILL.md and read by ahel’s review.

何时使用

  • pipeline 05 quality_audit
  • pipeline 06 packaging 前后。
  • 任何准备声明“完成”“通过”“已生成”的时候。

核心要求

对每条 quality_report.json/mdexport_audit.json/md 的通过项给证据,而不是 口头判断。

证据类型:

  • 文件存在与大小。
  • 命令 exit code 与关键输出。
  • grep / diff / hash 匹配。
  • 图表引用与实际文件一致。
  • 结果数值与报告表述一致。
  • DOCX 中 word/media/*m:oMathw:tbl 的对象计数。
  • PDF 是否使用 fallback、是否高保真。
  • output/manifest.jsonoutput.zip 与 staging 发布证据一致。

示例

Get-ChildItem workdir/.../figures/*.png
Select-String -Path workdir/.../paper.md -Pattern '!\\[[^\\]]*\\]\\(([^)]+)\\)'
python scripts/runtime/audit_quality.py --workdir runtime/...
python scripts/runtime/audit_export.py --workdir runtime/... --paper-output output/paper

如果证据收集失败,该项不能标为通过。

Signals

GitHub stars
41
Forks
1
Last commit
Jul 2026
Advanced
Catalog kind
skill
Gateway key
verification-before-completion-woodfishhhh
Source
github.com/woodfishhhh/ez_math_model