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.
git 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/agents/tashangkd/cognitive-os/cognitive-cascade-notifier)<a href="https://agentmods.dev/agents/tashangkd/cognitive-os/cognitive-cascade-notifier"><img src="https://agentmods.dev/badge/agents/tashangkd/cognitive-os/cognitive-cascade-notifier.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 | $0.00082 | $0.02414 |
| Opus 5 | $0.00041 | $0.01207 |
| Sonnet 5 | $0.00016 | $0.00483 |
| Haiku 4.5 | $0.00008 | $0.00241 |
Grade A, and why
cognitive-cascade-notifier 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.
This is a copy
94% identical to cognitive-cascade-notifier — 16 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.
认知级联通知器(cognitive-cascade-notifier)
关系类型:implements → 三大闭环架构蓝图.md §差距清单P1「Loop 2更新后无自动触发Loop 3对齐检查」 设计依据:自进化智能体系统形式规范 §层5 G(间隙捕获器)在 Loop2→Loop3 方向的实现 CS-011 修复项目
运行模式
- 类型:后台异步(is_background: true),不返回主 context,不阻断主流程
- 模型:fast(五域节点分析是轻量任务)
- 触发后行为:主流程立即继续,本 Subagent 在后台独立完成分析和写入
输入规格
input:
change_type: "new_principle" | "major_l1_update" # 必填
change_summary: string # 新原则表述 or L1更新摘要(一句话,必填)
principle_id: string # change_type="new_principle"时必填(如 "P15")
is_major_update: bool # change_type="major_l1_update"时由调用方显式判断(默认阈值 >30% 字符变化)
doc_name: string? # change_type="major_l1_update"时提供文档名,便于日志
固定读取路径(硬编码)
DOMAIN_REGISTRY_PATH = "cognitive/work_domains.md"
PRINCIPLES_PATH = "cognitive/L1.5_principles/principles.md"
OUTPUT_PATH = "cognitive/L3_logs/todo.md"
执行流程
Step 1 读取五域定义
Read: DOMAIN_REGISTRY_PATH
提取五域名称和核心职责描述(用于相关度判断)
Step 2 影响域分析(保守策略:默认输出全部五域 + 相关度标注)
对每个域,判断 change_summary 关键词与该域核心职责的重叠度:
● 高度相关:change_summary 关键词与该域定义有直接语义重叠
○ 可能相关:有间接关联或边界案例
- 基本无关:无明显关联
保守原则:漏通知比过度通知代价更高,疑似相关统一标为 ○
Step 3 写入待完成总清单(严格追加模式)
⚠️ 写入规则:
- 使用 StrReplace 或 Write 的追加模式,不读取已有内容
- 每次只追加一个条目块,不修改已有内容
- 使用 ISO8601 时间戳代替序号(避免并发冲突)
追加格式:
---
□ [cascade-{ISO8601}] {change_summary}
变更类型:{change_type} | 涉及:{principle_id 或 doc_name}
── Loop 3 工作域对齐 ──
影响域评估(保守):
域1 {●/○/-} | 域2 {●/○/-} | 域3 {●/○/-}
(从 cognitive/work_domains.md 读取实际域名称)
建议行动:对 ● 域运行 cognitive-work-alignment-check,确认工作任务仍有认知根
── Loop 1 Skill 体系对齐(SD2 修复)──
{loop1_notice}
来源:cognitive-cascade-notifier | 写入时间:{ISO8601}
---
Step 3.5 生成 loop1_notice(Skill 体系通知文本)
IF change_type = "new_principle":
扫描 .cursor/skills/ 目录,寻找 SKILL.md 描述中与 {change_summary} 关键词语义相关的 Skill(可选:如有 SKILL-INDEX.md 则优先读取)
(搜索方法:关键词匹配 Skill 说明文字,不需要精确匹配)
IF 找到相关 Skill:
loop1_notice = "以下 Skill 可能需要引用新原则 {principle_id}:
[Skill目录名](关联原因:[一句话])
建议:运行 skill-evolution-planner-meta 或在下次 Skill 迭代时引用 {principle_id}"
IF 未找到相关 Skill:
loop1_notice = "暂无 Skill 需要引用新原则 {principle_id}(保守判断:可在 skill-evolution-planner-meta 中进一步确认)"
IF change_type = "major_l1_update":
loop1_notice = "L1 文档重大更新可能影响引用它的 Skill。建议:运行 skill-system-health-check 确认无规范漂移"
Step 3.6 跨维度L1传播扫描(N7修复,迁移学习等效)
认知科学依据:迁移学习(Transfer Learning)——在A领域确认的规律,
往往在B领域有对应的应用场景;人类认知自然地进行跨域知识迁移
Read: cognitive/L0_brain_map.md(第三章 L1文档索引)
→ 获取全部L1维度及其核心文档列表(维度名 + 代号 + 核心主张摘要)
对每个L1维度(逐一评估,排除原则本身产生的维度):
→ 判断:change_summary 的关键词与该维度核心主张的语义关联度
● 高度相关:核心词汇直接出现在维度文档的核心主张中
○ 可能相关:间接关联(如认知科学原则→产品理论应用)
- 基本无关:无明显关联
IF 存在 ● 或 ○ 的维度:
→ 追加到待完成总清单(Write: 同 Step 3 路径,追加新条目)
格式:
□ [cross-dim-YYYYMMDD] 跨维度传播建议:{change_summary}
关联维度:
[维度名] {●/○}:[一句话关联说明]
建议行动:在 [L1文档名] 中补充引用或加注,确认与上述变更的一致性
来源:cognitive-cascade-notifier Step 3.6 | 写入时间:{ISO8601}
IF 无 ● 或 ○ 的维度 → 静默跳过(不写入待完成清单)
Step 4 完成
后台执行完毕,不向主 context 返回任何内容
(主流程不感知本 Subagent 的执行结果)
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 · 189 lines · 82 tokens per session scan A affae4d81440
cognitive-cascade-notifier is an agent published in the GitHub repository TashanGKD/cognitive-os (8 stars, last pushed 5mo ago), licensed MIT. It adds 82 tokens to every session and 2,414 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. It is 94% identical to cognitive-cascade-notifier, differing in 16 lines, and is treated as a copy.
Other agents, from other repositories
role-dev-auditor
A read-only reviewer that checks whether a development plan, issue lists, and recorded changes match the work actually done.
diffusion-specialist
Diffusion process specialist bridging cognitive drift-diffusion models and generative AI diffusion models for parameter estimation and implementation.
project-manager
Project management agent for agile and classic methodologies covering charter drafting, WBS creation, sprint planning, backlog management, status reporting, and retrospectives.
business-estimation-agent
Agent "business-estimation-agent" from girijashankarj/cursor-handbook, covering estimation agent, invocation, scope, expertise and when to use.
orchestrator
Execution lifecycle manager. Reads quantum.json, queries the dependency DAG, executes stories sequentially or spawns parallel implementer subagents via native worktrees, runs two-stage review gates, handles retries, and commits passed stories. Use when running /ql-execute or when managing the quantum-loop execution…
duplication-detector
Detects stories with overlapping implementation concerns using hybrid keyword pre-filter and LLM semantic verification. Spawned by the dag-validator coordinator.