counterexample-driven-plan-review
Skill2 starsChanged 11 days agoMITView on GitHub
反例驱动的只读方案审查。触发词:审方案、任务卡、判据反证、关键词假通过、同源证据冒充复核、固定轮数冒充收敛、计划被当成实施授权。用零写入纯函数探针构造正负反例,核对命令适用性、依赖闭包与验收层级,输出收敛后的任务卡。不用于代码实现或正式验收(verify-before-claim)。
In this repository131 other files it gives its agents.
- agent-client-network-triageSkill
- architecture-designerSkill
- async-python-patternsSkill
- audit-ci-cost-leversSkill
- awesome-design-mdSkill
- byok-model-retrySkill
- cloudbase-wechat-integrationSkill
- concurrent-edit-attribution-checkSkill
- continuation-state-auditSkill
- deliver-prdSkill
- desktop-update-push-triageSkill
- diagnose-github-actions-quotaSkill
- diagnose-windows-webview2-blank-windowSkill
- disk-junction-migrationSkill
- distinctive-design-systemsSkill
- dsh-session-format-debugSkill
- electron-app-stuck-spinner-triageSkill
- electron-asar-code-forensicsSkill
- electron-tray-icon-diagnosisSkill
- encrypted-office-visual-inspectionSkill
- frontend-design-ui-uxSkill
- frontend-developmentSkill
- game-designer-toolkitSkill
- game-numeric-designSkill
- github-acceleratorSkill
- global-config-smoke-runnerSkill
- goal-driven-executionSkill
- hid-device-protocol-probeSkill
- i-have-adhdSkill
- injected-context-slimmingSkill
- java-architectSkill
- java-qualitySkill
- kafka-best-practicesSkill
- large-sqlite-text-extractionSkill
- local-api-contractsSkill
- local-browser-e2eSkill
- local-db-migrationsSkill
- local-frontend-a11ySkill
- local-performance-evidenceSkill
- local-reverse-proxy-triageSkill
- local-rust-engineeringSkill
- local-spring-mybatisSkill
- mechanize-rule-into-hook-guardSkill
- memory-bank-divergence-repairSkill
- memory-entry-updateSkill
- merge-inbound-qoder-packageSkill
- minigame-dev-proSkill
- miniprogram-developmentSkill
- mongodb-atlas-stream-processingSkill
- mongodb-connectionSkill
- mongodb-mcp-setupSkill
- mongodb-natural-language-queryingSkill
- mongodb-query-optimizerSkill
- mongodb-schema-designSkill
- mongodb-search-and-aiSkill
- mysqlSkill
- nextjs-app-router-patternsSkill
- nginx-config-builderSkill
- nodejs-backend-patternsSkill
- oauth-login-failure-triageSkill
- The first 60 shown.
The file
Files
反例驱动的只读方案审查
适用
用户给来一份方案 / 任务卡 / 判据清单,要求“先审再实施”。目标:在不写文件的前提下,证明每条判据真的能失败、命令真的适用、验收层级没有被偷换。
1. 先核对授权与入口
- 本轮是只读审阅还是已授权实施?只读时零写入(含备份、队列、记忆、checkpoint)。
- 找到方案声称改动的真实代码入口/配置项;关键词存在不等于逻辑存在。
- 记录方案的验收判据,逐条编号。
2. 零写入探针
用纯函数 / 内存夹具验证,不落文件:
- Node stdin 或
node -e,避免 heredoc 反斜杠在传输中被改写;若 heredoc 失败,先区分传输失败与产品失败,不要把它算成方案缺陷。 - 直接调用真实导出的 API 或带锚点校验的源码函数;不手写桩冒充真实实现。
- 同一份内存夹具分别喂给“当前版”与“HEAD 基线版”,用独立期望值比对。
- 探针自身要有负对照:喂一个必然失败的输入,确认断言会红。
- 结束时核对内容未变(文件哈希/
git status),证明零写入。
3. 逐条反证判据
对每条判据问:
- 它读的是被测系统,还是探针自己刚写进去的夹具?
- 去掉目标实现后它还会绿吗?会绿 ⇒ 假判据。
- 关键词/字段存在是否被当成行为通过?改成行为断言或显式
UNVERIFIED。 - 证据是否同源?同一份脚本既产生又核验,不能算独立复核。
- 验收层级是否被偷换?静态 / 模拟 / 运行时 / 产物 / 原生宿主要分别标注。
4. 命令适用性
- 在真实 shell 里裸跑关键命令,记录退出码;管道会吞退出码,用
${PIPESTATUS[0]}或不用管道。 - Windows / Git Bash 的参数改写、路径转换、编码与行尾问题先排除。
- 命令依赖是否闭合:脚本引用的
client-root、测试导出、索引前置是否在隔离环境里存在;缺失会产生“未复现”的假结果。
5. 收敛判据:按覆盖,不按轮数
- 固定轮数(如“三轮无发现即收敛”)不是收敛判据;按覆盖矩阵收敛:每条判据都有正例、反例、负对照与证据层级。
- 新发现要回填到覆盖表;没有新覆盖就停,不要为凑轮数重复跑同一手段。
- 报告“审计通过”不等于“验收完整”;原始验收要求要逐项对账(源码、安装副本、真实调用、授权来源、产物证据)。
6. 输出:收敛后的任务卡
- 保留的判据、删除的假判据、新增的反例;
- 每条判据的命令与退出码;
- 未验证项与原因;
- 明确“这是审阅结论,不是实施授权”;实施需另行批准。
边界
- 只读审阅不修改方案文件、不更新状态、不调用 checkpoint。
- 不把本轮内存探针结果当成宿主端验证或已落地改进。
- 反例本身也要能失败;不能用一个恒真的反例证明判据有问题。
相关
- 通用证据纪律:
verify-before-claim。 - 规则/技能冲突审计:
rules-conflict-audit。 - 方案文档成文:
vibecoding-doc。
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.

