ai-output-verify
Skill0 starsChanged 24 days agoMITView on GitHub
验收 AI 生成或修改的游戏代码/文档时使用，针对「假完成、报告失真、跨模块回归」三类失效给出识别方法与门禁清单
In this repository12 other files it gives its agents.
- asset-import-pipelineSkill
- bug-diagnosis-chainSkill
- codex-rules-verifySkill
- config-table-pipelineSkill
- gameplay-breakdownSkill
- handoff-docSkill
- spec-card-authoringSkill
- unity-build-matrixSkill
- unity-commandline-playtestSkill
- unity-editor-verifySkill
- unity-smoke-assertionsSkill
- visual-evidence-reviewSkill
The file
Files
AI 产出验收门禁
何时触发
- AI 说「已完成 / 已修复 / 测试全过」，需要判断真假
- 改一处怕坏一片（数值、联机同步、卡牌结算这类互相牵连的系统）
三类高频失效
| 失效 | 典型表现 | 对策 |
| 假完成 | 声称实现，实际按钮无响应、数据没落盘、边界崩 | 完成定义写成「可观察行为」，要求给出跑通证据（日志 / 测试输出 / 实机效果） |
| 报告失真 | 汇报「已修复 / 全过」，实测仍报错，或改了没声明的文件 | 不信自述，只信断言与实测；用 diff 核对改动范围 |
| 跨模块回归 | 调 A 坏 B（数值 ↔ 结算 ↔ 同步） | 改动前跑基线、改动后跑全量；断言覆盖关键链路 |
门禁清单
- 需求与验收标准已写成文字，且可判断对错
- 给 AI 的上下文是最小充分的，没有让它猜接口
- 完成定义是可观察行为，不是「已实现」这种自述
- 关键路径逐段审查过（结算 / 存档 / 同步 / 界面状态）
- 断言基线全绿，且覆盖了受影响模块
- 有可复现的验证证据（日志 / 测试输出 / 实机效果）
- 回归范围与改动范围匹配，不是只跑了一部分
- 这次的坑已沉淀成规则或清单，防止复发
边界
设计取舍、数值判断、最终交付责任不外包给 AI。AI 提高的是实现与验证的速度，不改变设计判断的门槛。
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
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.