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.
npx skills add TashanGKD/cognitive-os --skill cognitive-associategit clone --depth 1 https://github.com/TashanGKD/cognitive-osWrote 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.
[](https://agentmods.dev/skills/tashangkd/cognitive-os/cognitive-associate)<a href="https://agentmods.dev/skills/tashangkd/cognitive-os/cognitive-associate"><img src="https://agentmods.dev/badge/skills/tashangkd/cognitive-os/cognitive-associate.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00089 | $0.01778 |
| Opus 5 | $0.00044 | $0.00889 |
| Sonnet 5 | $0.00018 | $0.00356 |
| Haiku 4.5 | $0.00009 | $0.00178 |
Grade A, and why
cognitive-associate 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 148 lines — stays where its author put it; the contents beside it link to each section on GitHub.
cognitive-associate(联想 / 启动效应等效)
对应认知活动:启动效应(Priming)——给定一个概念,自动激活语义相邻的概念网络,找到你可能没想到的关联
认知五维坐标(COG-TAX):
- 意识程度:前意识→显性(启动是无意识的,结果呈现是显性的)
- 脑网络:DMN(自由联想)+ SN(相关性检测)
- 记忆系统:语义记忆(L1)的扩散激活
- 执行功能:转换(在不同概念节点之间跳转)
- 双系统:系统1(模式匹配,快速)
理论依据:Collins & Loftus (1975) 扩散激活理论;Meyer & Schvaneveldt (1971) 启动效应实验
实现定位:能力层工具函数(不是对话层Skill) 触发条件:不接受用户直接触发词;只被其他Skill内部调用
知识导航表
| 层级 | 文档 | 用途 |
|---|---|---|
| D0 运行时数据 | cognitive/knowledge_graph.md |
概念关系边(depends/extends/cross_ref/references) |
| D0 上游输入 | 由调用方 Skill 传入「概念X」 | 激活起点 |
知识图谱说明:
- 维护者:cognitive-update-knowledge(新增L1文档时可选更新)/ cognitive-integrate-fragments(新增关系边时可选更新)
- 建立方式:初始建立需运行一次「建立知识图谱」任务(手动),后续随L1文档变更逐步更新
- 若不存在:本Skill降级为「仅返回L0地图中的维度关系」(功能大幅受限但不报错)
接口规格
输入(由调用方传入):
concept_x: str # 需要激活邻域的概念名称(通常是 L1 文档节点名或碎片标题)
hop_limit: int # 最大跳数(默认=2,不超过3)
输出(返回给调用方):
neighbors: list[{
name: str, # 邻居节点名称
relation: str, # 关系类型(depends/extends/cross_ref/references)
summary: str, # 邻居文档首段或100字摘要
relevance_score: float # 关系强度评分(depends=1.0 > extends=0.8 > cross_ref=0.6 > references=0.4)
}]
激活后立即执行
Step 1 接收输入
从调用方获取:concept_x(概念名)、hop_limit(默认=2)
Step 2 第一跳:直接邻居
Read: 知识图谱文件
→ 找 concept_x 对应节点的全部关系边
→ 提取所有直接邻居(第1跳):{邻居名, 关系类型}
Step 3 关系强度排序
按以下优先级排序:
depends(依赖)> extends(扩展/继承)> cross_ref(交叉引用)> references(提及)
→ 选取前5个作为「优先邻居」
→ 若邻居 < 3个,且 hop_limit ≥ 2:对排名前3的邻居继续做第2跳(Step 4)
Step 4 第二跳(可选,若第一跳邻居 < 3 且 hop_limit ≥ 2)
对第一跳前3个邻居,重复 Step 2,获取其邻居
→ 合并去重(排除 concept_x 自身)
→ 与第一跳邻居合并,仍按关系强度排序,取Top-5
Step 5 读取邻居摘要
对 Top-5 优先邻居,每个读取:
- 若存在 L0 地图中的 core_claim 字段 → 直接使用(< 80字)
- 若不存在 → 读取对应 L1 文档的首段(最多100字)
Step 6 组装并返回结果
返回格式(传给调用方,不输出到对话):
```
发现的关联概念(Top-5):
1. [邻居名](关系:depends)— [摘要]
2. [邻居名](关系:extends)— [摘要]
...
```
⚠️ 若知识图谱文件不存在或 concept_x 不在图谱中:
→ 返回「知识图谱未建立或无对应节点,联想激活跳过」
→ 不中断调用方的执行流程
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.
- 7d ago First seen · 148 lines · 89 tokens per session scan A 18e10e85c3db
cognitive-associate is a skill published in the GitHub repository TashanGKD/cognitive-os (8 stars, last pushed 5mo ago), licensed MIT. It adds 89 tokens to every session and 1,778 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.
Other skills, from other repositories
document-pipeline
A writing workflow for producing documents, articles, reports, and papers through research, drafting, fact checking, review, illustration planning, and format conversion.
project-retrospective
A project review workflow that examines the skills used during a completed project and records improvements for future work.
bug-fix-loop-coordinator
A document-driven process for fixing software bugs in priority order and testing each fix. TDD, or test-driven development, means writing a failing test before changing the code.
cognitive-reorganize
A workflow for reorganising scattered documents into a complete personal knowledge structure.
project-backlog
A project backlog manager that stores mixed requests in a project to-do document and can later process them by priority.
skill-capture-closure
A closing and review workflow for examining completed work, fixing defects in Skills, rules, or agents, and recording useful lessons in knowledge documents.