agentleFS
marblue939marblue939/ai-workflow-skills/skills/visual-evidence-review/SKILL.md

visual-evidence-review

Skill0 starsChanged 24 days agoMITView on GitHub

需要让模型判读游戏截图(HUD、卡牌演出、布局错位、穿模)时使用:取证规范、判读提问方式、结论回写与误判防范

The file

Files
SKILL.md

截图证据判读

何时触发

  • 需要确认「画面里到底有没有/对不对」,纯文字日志看不到的信息
  • 需要把一次改动的视觉结果留档,供复查

取证规范

  1. 固定视角与分辨率:同一问题的前后对比必须同机位同分辨率,否则差异无法归因。
  2. 命名可检索:日期_批次_场景_状态,例如 1005_批12_结算_墓碑缺失。
  3. 一图一问题:不要把五个问题塞进一张全景图,裁到出问题的区域。
  4. 保留原始帧:压缩图只用于看,判定以原图为准。

判读提问方式

  • 问「图里有没有 X / X 在什么位置 / 颜色是否 Y」,不要问「看起来对不对」。
  • 需要计数就问「数一数有几个」,并要求逐个数出来。
  • 对不确定的判读,换一个问法再问一次;两次不一致就人工看。
  • 判读结论必须能被原图复核——模型说「没有墓碑」,截图里那块地面也要能自证。

结论回写

  • 把判读结论写回诊断/工单文档,标注「视觉判读」而非「实测确认」。
  • 若判读与用户描述冲突,以用户实机反馈优先,再补证据。

验收

  • 前后对比图同视角同分辨率
  • 每个结论能指到图上的具体区域
  • 视觉判读与实测/日志结论分开标注
  • 原始帧已存档

坑

  • 用截图代替逻辑验证:画面正常不代表状态机正常。
  • 判读模型对文字密集的 UI 会漏读,关键文字要单独放大再问。
  • 图上没有参照物时很难判断「大小对不对」,截取时带上已知尺寸的元素。

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.