Learn Cog - The Tutor That Explains Five Different Ways
The best tutors explain the same concept five different ways. CellCog does too.
#1 on DeepResearch Bench (Apr 2026) for reasoning depth — deep enough to break concepts into first principles — combined with multi-modal output for every learning style: diagrams, analogies, worked examples, practice problems, interactive explanations, and full study guides. Any subject, any level.
How to Use
For your first CellCog task in a session, read the cellcog skill for the full SDK reference — file handling, chat modes, timeouts, and more.
OpenClaw (fire-and-forget):
CODEBLOCK0
All agents except OpenClaw (blocks until done):
from cellcog import CellCogClient
client = CellCogClient(agent_provider="openclaw|cursor|claude-code|codex|...")
result = client.create_chat(
prompt="[your task prompt]",
task_label="my-task",
chat_mode="agent",
)
print(result["message"])
How Learn-Cog Helps
Concept Explanations
Understand anything:
- - Break It Down: "Explain quantum entanglement like I'm 10 years old"
- Multiple Angles: "Explain recursion using 3 different analogies"
- Deep Dives: "Give me a comprehensive explanation of how neural networks learn"
- Visual Learning: "Create a diagram explaining the water cycle"
Example prompt:
"Explain blockchain technology:
Level: Complete beginner, no tech background
Include:
- - Simple analogy to start
- How transactions work
- Why it's secure
- Real-world examples
- Common misconceptions
Use simple language, avoid jargon. Include a visual diagram."
Homework & Problem Solving
Work through problems:
- - Math Problems: "Solve this calculus problem and explain each step"
- Science Questions: "Help me understand this physics concept and solve these problems"
- Essay Help: "Help me structure an essay on the causes of World War I"
- Code Debugging: "Explain why my code isn't working and help me fix it"
Example prompt:
"Help me understand this math problem:
Problem: Find the derivative of f(x) = x³sin(x)
I know basic derivatives but I'm confused about the product rule.
Please:
- 1. Remind me of the product rule
- Apply it step by step
- Give me 2 similar problems to practice
- Show me how to check my answer"
Study Materials
Prepare for success:
- - Study Guides: "Create a study guide for AP Chemistry exam"
- Flashcards: "Generate 50 flashcards for Spanish vocabulary"
- Practice Tests: "Create a practice quiz on US History 1900-1950"
- Summary Notes: "Summarize Chapter 5 of my biology textbook"
- Cheat Sheets: "Create a one-page reference for Python syntax"
Example prompt:
"Create a comprehensive study guide for the AWS Solutions Architect exam:
Cover:
- - Key services and when to use them
- Networking concepts
- Security best practices
- Cost optimization strategies
Format: Clear sections, bullet points, diagrams where helpful
Include: Practice questions after each section"
Coding & Tech Learning
Level up your skills:
- - Language Learning: "Teach me Python from zero to building a web app"
- Code Review: "Review my code and explain how to improve it"
- Project Tutorials: "Walk me through building a REST API step by step"
- Concept Deep Dives: "Explain how Docker containers actually work"
Example prompt:
"Teach me React hooks:
My level: I know basic JavaScript and HTML/CSS, never used React
Structure:
- 1. What problem do hooks solve?
- useState with simple examples
- useEffect with practical use cases
- When to use which hook
- A mini-project putting it together
Include code examples I can run."
Language Learning
Master new languages:
- - Grammar Explanations: "Explain Japanese particles with examples"
- Conversation Practice: "Practice ordering food in French"
- Writing Feedback: "Check my Spanish essay and explain my mistakes"
- Vocabulary Building: "Teach me 20 essential business Chinese phrases"
Learning Styles
Tell CellCog how you learn best:
| Style | Ask For |
|---|
| Visual | Diagrams, charts, infographics |
| Examples |
Multiple worked examples, real-world applications |
|
Analogies | Comparisons to familiar concepts |
|
Step-by-Step | Detailed breakdowns, numbered procedures |
|
Big Picture | Overview first, then details |
|
Hands-On | Practice problems, projects |
Subjects
CellCog can help with virtually any subject:
STEM:
- - Mathematics (all levels through advanced calculus and beyond)
- Physics, Chemistry, Biology
- Computer Science and Programming
- Statistics and Data Science
- Engineering concepts
Humanities:
- - History and Social Studies
- Literature and Writing
- Philosophy
- Languages
- Psychology
Professional:
- - Business and Finance
- Marketing
- Project Management
- Design
- Legal concepts
Tech Skills:
- - Programming languages
- Cloud platforms (AWS, GCP, Azure)
- DevOps and infrastructure
- Data engineering
- AI/ML concepts
Chat Mode for Learning
| Scenario | Recommended Mode |
|---|
| Homework help, concept explanations, practice problems | INLINECODE0 |
| Comprehensive study guides, full curriculum design, deep research |
"agent team" |
Use "agent" for most learning. Quick explanations, homework help, and study materials execute well in agent mode.
Use "agent team" for comprehensive learning - full course outlines, research papers, or when you need multi-source synthesis.
Example Prompts
Concept explanation:
"Explain the concept of recursion in programming:
My level: Beginner programmer, comfortable with loops
I need:
- - Clear definition
- Visual representation
- 3 progressively harder examples (factorial, fibonacci, tree traversal)
- Common mistakes to avoid
- When to use recursion vs iteration
Language: Python"
Exam prep:
"Create a study plan for the GRE:
Timeline: 2 months
Goal: 320+ score
Weak areas: Vocabulary and geometry
Include:
- - Weekly schedule
- Resources to use
- Practice test strategy
- Day-before checklist
Make it realistic for someone working full-time."
Language practice:
"Help me practice Japanese:
Level: JLPT N4
Focus: Conversational situations
Create a dialogue practice:
- - Scenario: Asking for directions in Tokyo
- Include vocabulary list
- Grammar points used
- Cultural notes
- Variations to practice
Use romaji and kanji with hiragana readings."
Tips for Better Learning
- 1. State your level: "Complete beginner" vs "I understand the basics" changes everything.
- 2. Ask why: Don't just ask for answers. Ask for explanations of the reasoning.
- 3. Request practice: Learning happens through doing. Ask for practice problems.
- 4. Admit confusion: "I don't understand the part where..." helps CellCog target explanations.
- 5. Build on previous: Reference what you already understand to get appropriate explanations.
- 6. Active recall: Ask CellCog to quiz you, not just explain. Testing improves retention.
If CellCog is not installed
Run /cellcog-setup (or /cellcog:cellcog-setup depending on your tool) to install and authenticate.
OpenClaw users: Run clawhub install cellcog instead.
Manual setup: pip install -U cellcog and set CELLCOG_API_KEY. See the cellcog skill for SDK reference.
Learn Cog - 五种不同方式讲解的导师
最优秀的导师会用五种不同方式讲解同一个概念。 CellCog 也是如此。
在DeepResearch基准测试(2026年4月)中排名第一,推理深度足以将概念分解为基本原理,同时为每种学习风格提供多模态输出:图表、类比、示例、练习题、互动讲解和完整学习指南。任何学科,任何水平。
如何使用
在会话中首次使用CellCog任务时,请阅读cellcog技能以获取完整的SDK参考——文件处理、聊天模式、超时等。
OpenClaw(即发即忘):
python
result = client.create_chat(
prompt=[您的任务提示],
notifysessionkey=agent:main:main,
task_label=my-task,
chat_mode=agent,
)
除OpenClaw外的所有代理(阻塞直到完成):
python
from cellcog import CellCogClient
client = CellCogClient(agent_provider=openclaw|cursor|claude-code|codex|...)
result = client.create_chat(
prompt=[您的任务提示],
task_label=my-task,
chat_mode=agent,
)
print(result[message])
Learn-Cog如何帮助
概念讲解
理解任何事物:
- - 分解讲解:像对10岁孩子一样解释量子纠缠
- 多角度讲解:用3个不同的类比解释递归
- 深入探讨:全面解释神经网络如何学习
- 视觉学习:创建解释水循环的图表
示例提示:
解释区块链技术:
水平:完全初学者,无技术背景
包括:
- - 从简单类比开始
- 交易如何运作
- 为什么安全
- 实际案例
- 常见误解
使用简单语言,避免专业术语。包括视觉图表。
作业与问题解决
解决问题:
- - 数学问题:解决这个微积分问题并解释每一步
- 科学问题:帮助我理解这个物理概念并解决这些问题
- 论文帮助:帮我构思一篇关于第一次世界大战原因的论文
- 代码调试:解释我的代码为什么不行并帮我修复
示例提示:
帮我理解这个数学问题:
问题:求f(x) = x³sin(x)的导数
我了解基本导数,但对乘积法则感到困惑。
请:
- 1. 提醒我乘积法则
- 逐步应用它
- 给我2个类似问题练习
- 展示如何检查答案
学习材料
为成功做准备:
- - 学习指南:为AP化学考试创建学习指南
- 闪卡:生成50张西班牙语词汇闪卡
- 模拟测试:创建美国历史1900-1950的练习测验
- 总结笔记:总结我的生物学教材第5章
- 速查表:为Python语法创建一页参考
示例提示:
为AWS解决方案架构师考试创建全面的学习指南:
涵盖:
- - 关键服务及其使用场景
- 网络概念
- 安全最佳实践
- 成本优化策略
格式:清晰的章节、要点、有用的图表
包括:每节后的练习题
编程与技术学习
提升技能:
- - 语言学习:从零开始教我Python到构建Web应用
- 代码审查:审查我的代码并解释如何改进
- 项目教程:逐步指导我构建REST API
- 概念深入:解释Docker容器实际如何工作
示例提示:
教我React hooks:
我的水平:了解基础JavaScript和HTML/CSS,从未使用过React
结构:
- 1. hooks解决了什么问题?
- 用简单示例讲解useState
- 用实际用例讲解useEffect
- 何时使用哪个hook
- 一个综合的小项目
包括可运行的代码示例。
语言学习
掌握新语言:
- - 语法讲解:用示例解释日语助词
- 对话练习:用法语练习点餐
- 写作反馈:检查我的西班牙语作文并解释错误
- 词汇积累:教我20个必备商务中文短语
学习风格
告诉CellCog你最适合的学习方式:
多个示例、实际应用 |
|
类比型 | 与熟悉概念比较 |
|
逐步型 | 详细分解、编号步骤 |
|
全局型 | 先概述,后细节 |
|
实践型 | 练习题、项目 |
学科
CellCog几乎可以协助任何学科:
STEM:
- - 数学(从基础到高等微积分及以上)
- 物理、化学、生物
- 计算机科学与编程
- 统计学与数据科学
- 工程概念
人文学科:
专业领域:
技术技能:
- - 编程语言
- 云平台(AWS、GCP、Azure)
- DevOps与基础设施
- 数据工程
- AI/ML概念
学习聊天模式
| 场景 | 推荐模式 |
|---|
| 作业帮助、概念讲解、练习题 | agent |
| 全面学习指南、完整课程设计、深度研究 |
agent team |
大多数学习场景使用agent。 快速讲解、作业帮助和学习材料在代理模式下执行良好。
全面学习使用agent team - 完整课程大纲、研究论文,或需要多源综合时。
示例提示
概念讲解:
解释编程中的递归概念:
我的水平:初级程序员,熟悉循环
我需要:
- - 清晰的定义
- 视觉表示
- 3个逐步变难的示例(阶乘、斐波那契、树遍历)
- 常见错误避免
- 何时使用递归vs迭代
语言:Python
考试准备:
为GRE创建学习计划:
时间线:2个月
目标:320+分
薄弱环节:词汇和几何
包括:
使其适合全职工作的人。
语言练习:
帮我练习日语:
水平:JLPT N4
重点:会话场景
创建对话练习:
- - 场景:在东京问路
- 包括词汇表
- 使用的语法点
- 文化注释
- 练习变体
使用罗马字和带平假名读法的汉字。
更好学习的技巧
- 1. 说明你的水平:完全初学者vs我了解基础会改变一切。
- 2. 问为什么:不要只问答案。要求解释推理过程。
- 3. 要求练习:学习通过实践发生。要求练习题。
- 4. 承认困惑:我不理解...的部分帮助CellCog精准讲解。
- 5. 基于已有知识:引用你已经理解的内容以获得合适的讲解。
- 6. 主动回忆:让CellCog考你,而不仅仅是讲解。测试提高记忆保持率。
如果未安装CellCog
运行/cellcog-setup(或根据工具运行/cellcog:cellcog-setup)进行安装和认证。
OpenClaw用户: 改为运行clawhub install cellcog。
手动安装: pip install -U cellcog并设置CELLCOGAPIKEY。参见cellcog技能获取SDK参考。