cognitive-extract-principle

cognitive-extract-principle is a skill for Cursor from TashanGKD/cognitive-os. It costs 78 tokens per session (4,728 once invoked), scanned A, original, MIT.

A Chinese-language skill for finding recurring patterns across separate notes and turning them into candidate underlying principles.

In plain words
What is it for?
Use it to review at least three related fragments, compare them with existing principles, detect conflicts, and propose a new principle for confirmation.
Why use it?
It helps organize scattered observations without treating a single example as a general rule. A person must approve a candidate before it is added to the principle library.

Skill for Cursor

Written for Cursor: installed under .cursor/.

Good fit Use it to review at least three related fragments, compare them with existing principles, detect conflicts, and propose a new principle for confirmation.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/tashangkd/cognitive-os/cognitive-extract-principle
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add TashanGKD/cognitive-os --skill cognitive-extract-principle
Clone the repo
git clone --depth 1 https://github.com/TashanGKD/cognitive-os

Made for: Cursor.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for cognitive-extract-principle

README.md
[![agentmods](https://agentmods.dev/badge/skills/tashangkd/cognitive-os/cognitive-extract-principle/github.svg)](https://agentmods.dev/skills/tashangkd/cognitive-os/cognitive-extract-principle)
Your own site
<a href="https://agentmods.dev/skills/tashangkd/cognitive-os/cognitive-extract-principle"><img src="https://agentmods.dev/badge/skills/tashangkd/cognitive-os/cognitive-extract-principle/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for cognitive-extract-principle

Your own site · 80×15
<a href="https://agentmods.dev/skills/tashangkd/cognitive-os/cognitive-extract-principle"><img src="https://agentmods.dev/badge/skills/tashangkd/cognitive-os/cognitive-extract-principle.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 78 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,728 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00078 $0.04728
Opus 5 $0.00039 $0.02364
Sonnet 5 $0.00016 $0.00946
Haiku 4.5 $0.00008 $0.00473

Measured 10d ago against content hash e6800c8e2463, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

cognitive-extract-principle scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.cursor/skills/cognitive-extract-principle/SKILL.md · 283 lines

How it starts

The opening of the file, as written. The whole thing — 283 lines — stays where its author put it; the contents beside it link to each section on GitHub.

L1.5原则提炼 Skill(Extract Principle)

实现「DMN模拟」:从多个L2碎片中发现跨领域共同模式,提炼为L1.5候选原则,经用户确认后正式沉淀。


知识导航表(执行前必须理解的概念根)

层级 文档 需要理解的概念
D0 认知根(必读) cognitive/L1.5_principles/principles.md K0原则的定义、现有原则列表(P1~Pn)、原则的稳定性要求(几乎不变)
D3 规范参考 cognitive/maintenance_protocol.md 原则确认流程:候选→用户确认→正式写入的三步要求
D4 运行时数据 cognitive/L2_fragments/碎片整合索引.md 所有🔲待整合碎片(提炼原则的原材料)

核心概念速查: ① K0原则 = 最高稳定性的K-object,跨所有场景成立,几乎不变——修改须谨慎 ② 提炼模式:≥3个碎片中出现相同底层模式,才候选升格为原则(避免过度泛化) ③ 用户确认是K0 ceremony的必要步骤:不允许AI自行写入原则库,必须等用户明确确认


激活后立即执行

Step 1  读取碎片数据
        Read: cognitive/L2_fragments/碎片整合索引.md(全部条目)
        Read: 全部L2碎片文件内容(按类型分批读取)
        Read: cognitive/L1.5_principles/principles.md(已有原则,避免重复提炼)

Step 2  分析跨碎片共同模式
        问题一:哪些碎片背后有相同的驱动逻辑?
        → 比较碎片的核心观点,找出「即使话题不同,但背后遵循同一个规律」的碎片群
        → 最少需要3个独立领域的碎片支持,才算可能成立的原则

        问题二:这个模式是否已被现有L1.5原则覆盖?
        → 与P1(验证优先于感受)、P2(从小点切入升维)对比
        → 如果是已有原则的印证 → 记录为「印证了P?」,不新建原则
        → 如果是新模式 → 继续

Step 2.5  [候选原则冲突检测](T4B 新增:在提出候选前检测与现有候选原则的冲突)
        快速扫描 底层原则库.md §二(候选原则列表 P4?~P17? 等):
        → 新发现的模式,与哪个现有候选原则在「某类场景下建议方向相反」?
        → 若发现冲突:
          「新发现的模式与候选原则 P??「[表述]」在以下场景存在优先级张力:
           [冲突场景描述]
           建议:在候选提出时明确说明两者的优先级关系。」
        → 将冲突关系纳入 Step 3 的候选展示中(供用户参考)
        → 若无冲突:静默通过,直接进入 Step 3

Step 3【γ】 提出候选原则(L1.5 原则是最高层认知,必须用户明确确认才能写入)
        对每个发现的新模式,生成:
        「━━ 候选原则 ━━
          候选表述:「[跨领域通用的原则,一句话]」

          印证碎片(N 个,跨 M 个独立领域):
          ┌─────────────────────────────────┐
          │ F-XXX(领域:产品理论)          │
          │ 「碎片摘要...」                  │
          │ F-YYY(领域:自我反思)          │
          │ 「碎片摘要...」                  │
          │ F-ZZZ(领域:组织设计)          │
          │ 「碎片摘要...」                  │
          └─────────────────────────────────┘

          关卡C:跨3个以上独立领域?[✅是 / ❌否]
          置信度:[高 | 中 | 低](基于印证数量和清晰程度)

          [✅ 确认为新原则,写入L1.5] [🔧 修改表述后确认] [❌ 不确认,继续观察]」
        → 等用户明确选择,不得在用户未确认前写入 L1.5

Step 3.5  【F-022 全节点挑战者反思】候选原则提出后、用户确认写入前执行
          以「原则体系批判者」视角执行3条挑战(在候选展示中附上挑战结论,供用户参考):
          
          1. 是否真的新:这条候选原则与 P1「验证优先于感受」和 P2「从小点切入升维」的本质区别
             是什么?能不能用一句话证明它不是P1或P2的特例?
          2. 反例构造:能否构造一个明确的反例——在某个场景下,遵循这条原则反而会
             导致更差的结果?如果很容易构造反例,候选原则的适用范围需要限定。
          3. 操作化检验:这条原则能否转化为一个「可以判断是否遵循了该原则」的具体行为描述?
             如果无法操作化,它可能还是一个模糊的价值观而非可执行的心智模型。
          
          挑战结论附在候选原则展示中:「挑战视角:[3条挑战的简短结论]」
          用户在看到候选原则时,同时看到挑战视角,再决定确认/修改/否定。

Step 4【α】 执行确认的新原则写入(用户已在 Step 3 授权,AI 直接写入所有后续步骤)
        用户确认后:
        a. Write:追加到 L1.5/底层原则库.md(新原则章节,含状态=✅已确认)
        b. 更新碎片整合索引.md:为相关碎片的「L1.5原则」字段填写新原则ID
        c. 执行L1约束检查(见 Step 5)
        d. 追加 cognitive/L3_logs/system_log.md
        e. 【情节记忆绑定】追加 L1.5/底层原则库.md 变更记录(见下方格式)
        f. 【工作记忆刷新 — L0状态地图更新】更新 L0_大脑总地图.md:
           → 「最后更新」时间行:追加本次原则确认/新增的描述
           → §二 L1.5 底层原则层:更新已确认原则数(如有新原则升格)
           认知科学依据:Baddeley 工作记忆情境模型(Situational Model)——L0是整个认知
           体系的「当前情境表征」,每次图式变化后必须刷新以维持情境一致性
        g. 【语义记忆组织更新 — 知识图谱状态行】若本次原则确认触发了新的L1约束检查
           并导致L1文档新增节点,更新 知识图谱_正式文档.md:
           → 版本号递增(如 v1.8 → v1.9)
           → 节点统计行更新(「N篇L1 + M篇REF-EXT」)
           → 若未新增节点:跳过此步
           认知科学依据:Tulving 语义记忆组织(Semantic Memory Taxonomy Maintenance)——
           知识分类账(图谱)的状态头等价于语义记忆的「当前分类状态摘要」,
           任何分类变化后必须同步
           ⚠️ 这是 ceremony(K0) 的必要步骤:原则库作为最高稳定性K-object,
              每次结构性修改(新增/升格原则)必须留下情节记忆记录
           
           追加到原则库末尾的变更记录 §:
           ```
           ### YYYY-MM-DD — 确认新原则 PX「[表述摘要,10字以内]」(从候选升格)

Read the full file on GitHub · 283 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 283 lines · 78 tokens per session scan A e6800c8e2463

Subscribe to this mod's changes

cognitive-extract-principle is a skill published in the GitHub repository TashanGKD/cognitive-os (9 stars, last pushed 5mo ago), licensed MIT. It adds 78 tokens to every session and 4,728 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.