← 返回 Skills 市场
nitishgargiitd

Learn Cog

作者 CellCog · GitHub ↗ · v1.0.12 · MIT-0
darwinlinuxwindows ✓ 安全检测通过
3474
总下载
7
收藏
15
当前安装
13
版本数
在 OpenClaw 中安装
/install learn-cog
功能描述
AI tutoring and education powered by CellCog. Study guides, exam prep, coding tutorials, language learning, math help, science explanations, practice problem...
使用说明 (SKILL.md)

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):

result = client.create_chat(
    prompt="[your task prompt]",
    notify_session_key="agent:main:main",
    task_label="my-task",
    chat_mode="agent",
)

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
  2. Apply it step by step
  3. Give me 2 similar problems to practice
  4. 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?
  2. useState with simple examples
  3. useEffect with practical use cases
  4. When to use which hook
  5. 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 "agent"
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.

安全使用建议
This skill appears to be a plain adapter for CellCog's tutoring API. Before installing: 1) Confirm the CELLCOG_API_KEY comes from CellCog and you trust their data handling and privacy policy; any prompt content (including student answers or uploaded problems) will be sent to that external service. 2) Do not paste passwords, private keys, or other secrets into prompts. 3) Ensure python3 and the cellcog client are installed in the agent environment (the skill lists 'cellcog' as a dependency but provides no installer). 4) If you need offline-only tutoring or local-only code execution, this skill is not suitable because it relies on the CellCog cloud API.
功能分析
Type: OpenClaw Skill Name: learn-cog Version: 1.0.12 The learn-cog skill bundle is a documentation-focused package providing instructions and prompt examples for an AI tutoring service. It contains no executable code, and its SKILL.md content is entirely aligned with its stated purpose of educational assistance (math, coding, language learning). No indicators of data exfiltration, malicious execution, or prompt injection were found; it simply references a 'cellcog' Python dependency and API key required for its functionality.
能力标签
cryptorequires-sensitive-credentials
能力评估
Purpose & Capability
Name/description (AI tutoring via CellCog) match the declared requirement of CELLCOG_API_KEY and the examples showing use of the CellCog Python SDK. Requesting python3 and a CellCog API key is proportionate to a Python-based client that calls an external tutoring API.
Instruction Scope
SKILL.md contains instructions to call the CellCog API via the cellcog Python client (create_chat). It does not instruct reading unrelated local files, scanning system configs, or exfiltrating other credentials. Note: examples reference notify_session_key and agent session behavior — prompts sent to the API will be transmitted externally, so do not include secrets in prompts.
Install Mechanism
This is instruction-only (no install spec or code files), which is low-risk. The manifest lists a dependency 'cellcog' but provides no install steps; you must ensure python3 and the cellcog client are present in the environment. No downloads or arbitrary URLs are used.
Credentials
Only CELLCOG_API_KEY is required, which is appropriate for a third-party API client. No additional unrelated secrets or system config paths are requested.
Persistence & Privilege
The skill is not always-on and uses the platform defaults for invocation. It does not request persistent system modifications or access to other skills' configurations.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install learn-cog
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /learn-cog 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.12
- Added `requires` section to metadata, specifying dependency on `python3` and environment variable `CELLCOG_API_KEY`. - No changes to user-facing features or documentation content.
v1.0.11
- Updated the description to be more concise and to better reflect the skill’s features. - Clarified usage instructions for OpenClaw and other agent types. - Minor editorial refinements for clarity and consistency throughout the SKILL.md. - No functionality or user-facing feature changes; documentation only.
v1.0.10
- Updated description with more concise wording and enhanced focus on five-way explanations and DeepResearch ranking. - Improved and clarified SDK example usage, including explicit agent_provider setup for non-OpenClaw platforms. - Refined introductory and instructional language for easier onboarding. - No changes to features or functionality; documentation only.
v1.0.9
- Simplified and clarified skill description for brevity. - Added a "How to Use" section with updated quickstart instructions, emphasizing using the cellcog skill for full SDK details. - Consolidated and streamlined the explanations for usage in different agent environments. - Reformatted and focused sections for Concept Explanations, Homework & Problem Solving, Study Materials, Coding/Tech Learning, and Language Learning. - Cleaned up tips and added missing steps for getting started if CellCog is not installed. - Removed redundant and overly detailed sections to make the documentation more concise and accessible.
v1.0.8
- Expanded and restructured documentation with detailed examples and prompt templates for every subject and use-case. - Emphasized CellCog’s ability to explain concepts five different ways: diagrams, analogies, worked examples, interactive lessons, and more. - Included sample prompts for concept explanations, homework help, study materials, coding tutorials, and language practice to demonstrate versatility. - Updated subject coverage lists and added guidance on tailoring responses for different learning styles. - Detailed recommended chat modes for various learning scenarios. - Added new tips to help users get the most effective educational support.
v1.0.7
learn-cog 1.0.7 Changelog - Updated SKILL.md: streamlined content and clarified capabilities. - Shortened and simplified the description. - Reduced detailed examples and feature explanations; focused on summary of outputs and supported formats. - Added a new "Related Skills" section referencing research-cog and docs-cog. - Kept prerequisite and usage instructions, but removed lengthy subject and prompt samples for brevity.
v1.0.6
- Added specific OpenClaw usage instructions, including recommended "fire-and-forget" agent pattern and session key for long tasks. - Updated quick start and SDK usage examples for clarity on agent selection and blocking vs. non-blocking calls. - Streamlined SDK reference note, removing redundant details and pointing users to primary skill documentation. - No functional or interface changes to the tutoring/coaching workflows.
v1.0.5
- Updated the Prerequisites section with a simplified "Quick start" code example. - Referred users to the cellcog skill for detailed SDK usage, delivery modes, and advanced API features. - Removed fire-and-forget / notify session key code example in favor of standard usage. - No functional or interface changes; documentation improvement only.
v1.0.4
- Updated DeepResearch Bench ranking date in documentation from Feb 2026 to Apr 2026. - No functional or interface changes; documentation alignment only.
v1.0.3
- Updated skill description for clarity and broader appeal. - Added supported operating systems and homepage link to metadata. - No changes to core functionality; documentation improvements only.
v1.0.2
- Added author and dependencies fields to skill metadata for clarity and usability. - Updated prerequisite instructions: clarified that `cellcog` is now a required dependency for SDK setup and API calls. - No changes to core learning modes or features; documentation and usage patterns remain the same.
v1.0.1
- Enhanced focus on multi-modal, multi-approach explanations (diagrams, analogies, examples, practice, interactive) across all subjects. - Updated description and introduction to emphasize reasoning depth and ranking (#1 on DeepResearch Bench, Feb 2026). - Added highlight of “explaining concepts five different ways” as Learn-Cog’s core advantage. - Refined messaging throughout for clarity and stronger positioning as a top-tier AI tutor. - No changes to functionality, examples, or usage patterns; update is documentation-only.
v1.0.0
Learn-cog 1.0.0: Initial Release - Launches AI-powered learning assistant leveraging CellCog. - Offers support for homework help, tutoring, study guides, explanations, coding tutorials, language learning, and exam prep. - Provides example prompts for a variety of use cases and learning styles. - Requires installation and setup of the CellCog skill for API access. - Details use of "agent" and "agent team" chat modes for tailored learning experiences. - Comprehensive documentation for effective prompts and best practices.
元数据
Slug learn-cog
版本 1.0.12
许可证 MIT-0
累计安装 15
当前安装数 15
历史版本数 13
常见问题

Learn Cog 是什么?

AI tutoring and education powered by CellCog. Study guides, exam prep, coding tutorials, language learning, math help, science explanations, practice problem... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 3474 次。

如何安装 Learn Cog?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install learn-cog」即可一键安装,无需额外配置。

Learn Cog 是免费的吗?

是的,Learn Cog 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Learn Cog 支持哪些平台?

Learn Cog 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(darwin, linux, windows)。

谁开发了 Learn Cog?

由 CellCog(@nitishgargiitd)开发并维护,当前版本 v1.0.12。

💬 留言讨论