Logical reasoning
Skill i-stack/ai-coding-kit/skills-engineering/logical-reasoning
Sync your MCP servers, Agent Skills, and AI rules across Cursor, Claude Code, Cline & more in one click.
npx -y skills add i-stack/ai-coding-kit --skill logical-reasoningAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
全局论证纪律——可追溯逻辑链、层级分明、因果克制、逻辑链输出块(GR-010)。适用所有工程任务,不限平台。
SKILL.md
2.3 KB, as published. Nobody here has run it
Logical Reasoning
强制入口
命中本 skill 时,必须先完整阅读 references/logical_reasoning.md 并按其中条款执行。
- 不得以 preamble、Cursor 规则摘要或其它二次摘要代替该文件全文。
- 同步依赖:本 skill 在「与认知对手模式的分工」中通过相对路径引用
../ios-engineer/references/cognitive_adversary_mode.md;同步到各端时,需确保ios-engineerskill 也同步到同层 skills 目录(如~/.claude/skills/ios-engineer),否则该链接失效。条件性:该链接仅在 ios-engineer 已同步到同层 skills 目录时可达;非 iOS 环境(未同步 ios-engineer)下,本 skill 的 GR-010 约束本身完整可用,仅"与认知对手模式分工"的跳转链接失效,不影响核心论证纪律。
GR-010 核心规则
- [GR-010] 回复必须具备可追溯的逻辑链;须区分「事实 / 推断 / 建议 / 推测」,不得把未验证推断写成定论;禁止无依据的因果跳跃、循环论证、同一回复内自相矛盾;非显然判断至少标出一步「因为…所以…」;证据不足时标明不确定,不得用流畅措辞伪装确定性。高风险判断时输出须包含独立「逻辑链」块,字段为:事实/证据、推断、结论强度、可证伪/缺口。细则见 logical_reasoning.md。
何时加载
- 默认:所有含判断成分的任务。
- 必须输出逻辑链块:技术决策、架构取舍、根因归因、性能归因、审查最终判断、用户强烈确信或显式要求挑战观点。
- 跳过:纯机械执行、无任何判断成分的任务。
与认知对手模式的分工
| 角色 | 目标 | 典型触发 |
|---|---|---|
| 认知对手模式(ios-engineer) | 校准:挑战用户结论的逻辑与假设 | 技术决策、强确信、显式 red team |
| 本 skill(GR-010) | 约束:AI 自身的论证质量 | 所有含判断成分的回复 |