cognitive-capture-fragment

cognitive-capture-fragment is a skill for Cursor from TashanGKD/cognitive-os. It costs 58 tokens per session (2,639 once invoked), scanned A, original, MIT.

A note-capture workflow that turns quick thoughts into structured entries in a cognitive knowledge system.

In plain words
What is it for?
Use it to record insights, observations, and incomplete ideas, assign them to the right area, and update the system’s index and log.
Why use it?
It keeps fleeting ideas from being lost and checks whether a thought should add to an existing document or become a new note.

Skill for Cursor

Written for Cursor: installed under .cursor/.

Good fit Use it to record insights, observations, and incomplete ideas, assign them to the right area, and update the system’s index and log.

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

README.md
[![agentmods](https://agentmods.dev/badge/skills/tashangkd/cognitive-os/cognitive-capture-fragment/github.svg)](https://agentmods.dev/skills/tashangkd/cognitive-os/cognitive-capture-fragment)
Your own site
<a href="https://agentmods.dev/skills/tashangkd/cognitive-os/cognitive-capture-fragment"><img src="https://agentmods.dev/badge/skills/tashangkd/cognitive-os/cognitive-capture-fragment/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-capture-fragment

Your own site · 80×15
<a href="https://agentmods.dev/skills/tashangkd/cognitive-os/cognitive-capture-fragment"><img src="https://agentmods.dev/badge/skills/tashangkd/cognitive-os/cognitive-capture-fragment.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 58 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,639 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 $0.00058 $0.02639
Opus 5 $0.00029 $0.01319
Sonnet 5 $0.00012 $0.00528
Haiku 4.5 $0.00006 $0.00264

Measured 11d ago against content hash c334cdc8bc49, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

cognitive-capture-fragment 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 11d 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-capture-fragment/SKILL.md · 186 lines

How it starts

The opening of the file, as written. The whole thing — 186 lines — stays where its author put it; the contents beside it link to each section on GitHub.

碎片捕捉 Skill(Capture Fragment)

把用户随时产生的碎片想法,经过关卡A(新洞见判断)后,结构化写入L2碎片层,并更新索引和系统日志。


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_knowledge/formal_spec/self_evolving_agent_spec.md 层1:K-object定义(编码陈述性事实的知识对象);层2:D_op.create(对象应存在但不存在);层7:完备题库维度——对象类型 × 操作类型 × 关系类型
D3 规范参考 cognitive/maintenance_protocol.md 碎片写入规范:命名格式/存放位置/索引更新要求
D4 运行时数据 cognitive/L2_fragments/fragment_index.md + L0_大脑总地图.md 当前最大碎片ID(F-XXX)+ 整体认知体系地图(确定碎片归属域)

核心概念速查: ① K-object.create(捕捉碎片)= 系统中尚不存在的知识,需要新建一个K-object来编码它 ② L2碎片 = K-object的fragment子类,是L1框架文档的上游原材料,待整合 ③ 写入后必须更新碎片整合索引——实现C1(完全可观测性:所有对象必须在R中注册)


激活后立即执行(顺序不可跳过)

Step 1  读取上下文
        Read: cognitive/L2_fragments/fragment_index.md
        → 获取当前最大碎片ID(F-XXX),下一个ID = F-(N+1)
        Read: cognitive/L0_brain_map.md(摘要,了解当前知识体系)
        Read: cognitive/document_catalog.md(快速扫描,识别与当前话题相关的已有文档)
        → 若发现分类清单中有与当前碎片主题高度相关的已有文档(★CURRENT 或 ⏸️DEFERRED):
          提示用户「已有相关文档:[文档名] / [Loop归属] / [状态],是补充该文档还是记录新碎片?」
          → 用户选"补充已有文档" → 结束本 Skill,引导至 cognitive-update-knowledge
          → 用户选"记录新碎片" → 继续 Step 2

Step 2  理解用户输入
        → 用户说的是完整想法,还是需要追问才能清楚?
        → 如果不够清楚,追问一个最关键的问题再继续(不要追问多个)

Step 3  [关卡A] 判断是否有新洞见
        → 快速比对碎片整合索引中的已有标题,判断是否与已有内容高度重复
        → 如果高度重复(语义相似度>80%):
          「此内容与碎片[F-XXX]「...标题...」高度相似。是追加补注,还是确实是新视角?」
          → 用户说"追加"  → 在原碎片文件对应条目末尾追加注释,更新索引,写日志,结束
          → 用户说"新的"  → 继续 Step 4
        → 如果是新洞见:继续 Step 4

Step 4  推断碎片属性
        根据内容推断:
        - 类型:product_theory(产品理论)| self_reflection(自我反思)| ai_review(AI复盘)
                org_design(组织设计)| methodology(思维方法)
        - 对应L1文档(初步关联):[A]产品框架 / [B]产品定位 / [B']产品设计 / [C]协作模式 /
                                  [D]元框架 / [E]文档体系 / 个人思维方法论 / 个人认知模式
        - 对应L1.5原则:P1(验证优先于感受)/ P2(从小点切入升维)/ 无
        - 确认程度:confirmed(确信)/ tentative(暂定)/ uncertain(存疑)

Step 5  生成结构化碎片并向用户确认
        展示:
        「准备记录碎片 [F-新ID]:
          标题:「...」
          类型:[类型]
          归因:🔵 用户原始思考
          关联L1:[X]
          关联L1.5:P? / 无
          确认程度:[确认程度]
          [确认记录] [修改后记录]」

Step 5  用户确认写入后,正式将碎片写入 L2 文件
        (按 cognitive-l3-auto-log 规则,写入后自动追加系统日志)

Step 5.5  【F-022 全节点挑战者反思】用户确认后、正式写入前执行
          以「认知体系一致性守门人」视角执行3条挑战:
          
          1. 矛盾检测:这条碎片的核心观点,与 L1.5 已确认原则(P1/P2)或 L1 文档中
             任何已有结论有没有张力(不一定是矛盾,但可能需要说明关系)?
          2. 置信度诚实:碎片中有没有「用确定语气说了一个实际上是推断的结论」?
             如果有,应该在碎片中标注「推断/待验证」而非写成已确认事实。
          3. 级联缺失:写入这条碎片后,是否有任何关联的 L1 文档应该被标注为「需重新审视」,
             但目前碎片的关联字段没有指向它?
          
          若发现可修复的问题(措辞/置信度/关联)→ 先修改碎片内容,再写入
          若确实无重大问题 → 继续写入

Step 5.8  L2 积累阈值监控(CS-014 修复,写入后检查,静默执行)
          
          Read: cognitive/L2_fragments/fragment_index.md(快速计数)
          统计 🔲 待整合 状态的碎片总数 N
          
          IF N ≥ 8:
            「💡 当前积累了 N 条待整合碎片。建议触发 cognitive-integrate-fragments 处理。
               [立即整合] [稍后处理]」
          IF N < 8:静默,不输出任何提示

Step 6  执行写入(用户确认后)
        a. Write 到对应的L2文件(追加到文件末尾):
           - product_theory/self_reflection/ai_review/org_design/methodology
           → 文件路径:cognitive/L2_fragments/[类型目录]/[文件名].md
        b. 更新碎片整合索引.md(追加一行:ID | 标题 | 类型 | 关联L1 | L1.5原则 | 🔲待整合 | 时间)
        c. 追加 cognitive/L3_logs/system_log.md:
           [LOG-今日日期-NN] cognitive-capture-fragment | 记录碎片[F-ID]「标题」 | L2碎片文件+索引

Step 7  收尾反馈
        「✅ 碎片 [F-ID] 已记录。
          当前待整合碎片:N 个
          [现在整合最近的碎片] [稍后处理]」
        → 如果用户选择"现在整合",触发 cognitive-integrate-fragments Skill

Read the full file on GitHub · 186 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. 11d ago First seen · 186 lines · 58 tokens per session scan A c334cdc8bc49

Subscribe to this mod's changes

cognitive-capture-fragment is a skill published in the GitHub repository TashanGKD/cognitive-os (9 stars, last pushed 5mo ago), licensed MIT. It adds 58 tokens to every session and 2,639 once invoked, about $0.0003 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.