MCPcopy Index your code
hub / github.com/datawhalechina/hello-claw

github.com/datawhalechina/hello-claw @v1.0.0

Chat with this repo
repository ↗ · DeepWiki ↗ · release v1.0.0 ↗ · + Follow
47 symbols 140 edges 7 files 0 documented · 0%
What it actually does AI analysis from the code graph — generated when you open this
loading…
README

Hello Claw Logo

In an unremarkable repository, a lobster was born.

Pick one and send it to school; or write your own undefined lobster.

Its dreams have been big from day one.

Hello Claw 👋

Hello Claw: adopt your AI lobster assistant, learn Skills in Lobster University, and build your own intelligent agent from scratch

📌 Read Online | 💬 Join Community | 🚀 Want to Learn Vibe Coding Too?

<a href="https://github.com/datawhalechina/hello-claw/stargazers" target="_blank">
    <img src="https://img.shields.io/github/stars/datawhalechina/hello-claw?color=660874&style=for-the-badge&logo=star&logoColor=white&labelColor=1a1a2e" alt="Stars"></a>
<a href="https://github.com/datawhalechina/hello-claw/network/members" target="_blank">
    <img src="https://img.shields.io/github/forks/datawhalechina/hello-claw?color=660874&style=for-the-badge&logo=git-fork&logoColor=white&labelColor=1a1a2e" alt="Forks"></a>
<a href="https://github.com/datawhalechina/hello-claw/raw/v1.0.0/LICENSE" target="_blank">
    <img src="https://img.shields.io/badge/License-CC_BY_NC_SA_4.0-4ecdc4?style=for-the-badge&logo=creative-commons&logoColor=white&labelColor=1a1a2e" alt="License"></a>

简体中文 English 日本語

[!WARNING] 🧪 Beta Notice: the main tutorial is complete, and we are now polishing the details. Issues and suggestions are very welcome.

Read Online

https://datawhalechina.github.io/hello-claw

Introduction

This project is a complete learning guide for OpenClaw, designed to help you master this powerful command-line AI assistant system from zero. Whether you want to use OpenClaw quickly to boost your productivity or deeply understand how it works and build your own version, this tutorial gives you a clear path forward.

This project contains three core modules:

  1. Adopt Claw (User Guide): 11 chapters plus 7 appendices, covering installation (Ch. 1-3), core configuration (Ch. 4-6), operations and extensions (Ch. 7-9), and security plus client interfaces (Ch. 10-11)
  2. Lobster University (Practical Scenarios): hands-on case studies centered on Skills selection and reusable workflow patterns
  3. Build Claw (Developer Guide): 15 chapters that first break down OpenClaw source code and alternatives, then move into Skills, channels, and full customization

Who is this for:

  • Beginners: want an always-available AI assistant without needing prior programming experience
  • Productivity enthusiasts: want to control AI remotely through QQ, Feishu, or Telegram
  • Technical explorers: are curious about OpenClaw's skill system and automation capabilities
  • Developers: want to understand agent architecture deeply and build their own version

Suggested learning paths:

  • Beginners: start with Part 1 "Adopt Claw" and get installation plus basic automation working first
  • If you want scenario-driven outcomes fast: jump into "Lobster University" and pick 5-10 Skills by use case
  • Developers: go to "Build Claw" to study the underlying implementation and customize your own Claw

🔥 News

  • [2026-03-12] ✅ Completed Build Claw Ch. 1-10: core architecture analysis (prompt system, tool system, message loop, multi-channel integration), alternative solution exploration (lightweight, hardened security, hardware), and a summit-view retrospective
  • [2026-03-10] ✅ Completed Build Claw Ch. 13: Skill file structure, frontmatter, async handling, and debugging
  • [2026-03-10] ✅ Added Lobster University: a menu-style Skills elective guide to give your lobster "combat add-ons"
  • [2026-03-08] ✅ Completed Adopt Claw Ch. 1-11: installation (AutoClaw + manual install + onboarding wizard), core configuration (chat platforms, models, agents), operations and extensions (tools and scheduled tasks, gateway, remote access), and security plus clients (security protection, web interface)
  • [2026-03-04] 🦞 Project launched with two planned core modules: "Adopt Claw" and "Build Claw"

📖 Table of Contents

Part 1: Adopt Claw (User Guide, 11 Chapters + Appendices A-G)

Chapter Description Status
Opening Notes What OpenClaw is, the four-step adoption method, and the learning roadmap
🔵 Installation
Ch. 1 One-Click AutoClaw Installation Download the AutoClaw desktop client and get a zero-barrier experience in 5 minutes
Ch. 2 Manual OpenClaw Installation Terminal basics, Node.js installation, npm install, and the onboarding wizard
Ch. 3 Initial Configuration Wizard CLI wizard, macOS guidance, custom providers, and reconfiguration
🟢 Core Configuration
Ch. 4 Chat Platform Integration Platform overview, full Feishu integration example, pairing, and group chats
Ch. 5 Model Management Model concepts, CLI management, multi-provider setup, API key rotation, and failover
Ch. 6 Agent Management Multi-agent management, workspaces, heartbeat, and binding rules
🟡 Operations and Extensions
Ch. 7 Tools and Scheduled Tasks Toolset levels, scheduled tasks (cron / at / every), and web search
Ch. 8 Gateway Operations Startup management, hot reload, authentication security, key management, sandbox policy, and log monitoring
Ch. 9 Remote Access and Networking SSH tunnels, Tailscale networking, deployment architecture, and security best practices
🔴 Security and Clients
Ch. 10 Security Protection and Threat Model Threat landscape, VM isolation, trust boundaries, MITRE ATLAS, and supply-chain security
Ch. 11 Web Interface and Clients Dashboard, WebChat, Control UI, TUI, and third-party clients
Appendices
Appendix A: Learning Resources 8 categories of resources, 80+ links, carefully curated by the editors
Appendix B: Community Voices and Ecosystem Outlook Deep discussion of 6 major topics plus highlighted quotes
Appendix C: Claw-Like Solution Comparison and Selection 5 major categories: desktop clients, hosted services, cloud vendors, self-hosted open source, and mobile
Appendix D: Skill Development and Publishing Guide SKILL.md format, skill-creator, and the ClawHub publishing workflow
Appendix E: Model Provider Selection Guide System comparison across aggregators, mainland China, international, and local providers
Appendix F: Command Quick Reference Reference for all CLI commands covering installation, configuration, logs, cron, channels, and more
Appendix G: Configuration File Deep Dive Detailed explanation of every field in openclaw.json

Part 2: Build Claw (Developer Guide, 15 Chapters)

Chapter Description Status
Opening Notes Why build your Claw from scratch, the OpenClaw complexity challenge, and the learning roadmap
🔵 OpenClaw Internal Breakdown (Ch. 1-7)
Ch. 1 Architecture Design Philosophy The evolution of AI agent architecture and OpenClaw's six architectural innovations
Ch. 2 ReAct Loop The agent's "think-act" engine and execution loop
Ch. 3 Prompt System Prompt architecture, hot reload mechanism, and persistent persona design
Ch. 4 Tool System A deep dive into the four primitive tools, registration, and composition capabilities
Ch. 5 Message Loop and Event-Driven Design Swimlane model, heartbeat mechanism, concurrency safety, and time initiative
Ch. 6 Unified Gateway Gateway architecture, multi-channel integration, and message normalization
Ch. 7 Security Sandbox Balancing freedom and constraints, execution isolation, and permission control
🟢 Custom Solution Paths (Ch. 8-10)
Ch. 8 Lightweight Solutions Community variants such as NanoClaw, Nanobot, and ZeroClaw
Ch. 9 Security Hardening Solutions IronClaw security architecture, sandbox isolation, and audit logs
Ch. 10 Hardware Solutions PicoClaw hardware selection and low-power embedded deployment
🟡 Level 3: Customize Your Claw (Ch. 11-15)
Ch. 13 Skill Authoring Skill file structure, frontmatter, async handling, and debugging

Part 3: Lobster University (Practical Scenarios)

Article Description Status
Lobster University: Opening Notes and Skills Elective Guide Dual entry through ClawHub and SkillHub, skill taxonomy, course-selection principles, and recommended schedules
Vibe Coding in Practice A full loop from Feishu conversation to requirements and PR: "conversation as development"
Email Assistant in Practice (163) From IMAP/SMTP setup to script validation and cron-based email automation
Multi-Agent Collaboration (Multi OpenClaw / HiClaw) From a single-worker bottleneck to multi-agent collaboration with traceable task orchestration
Security Checklist Security boundaries for skills and tools, plus auditing and anti-injection essentials
Skill Development in Practice: Local Health Assistant Build a local-first home health management loop powered by MediWise
---

🎉 We'd love case-study contributions!

If you have a unique OpenClaw scenario or practical experience to share, you are very welcome to contribute in any of these ways:

  • Submit a PR to add your case study to this section
  • Open an Issue describing your scenario
  • Join the community discussion and exchange ideas with other developers

Every contribution helps more people discover what OpenClaw can do.

🦞 Use Case Gallery (Continuously Updated)

🌅 Personal Productivity • Morning briefing (weather + schedule + todos) • Automatic email classification and summaries • Smart calendar management 💻 Programming and Development • Code generation and review • Automated testing and deployment • Automatic documentation generation 📢 Content Creation • Automated social media operations • Writing assistance and polishing • Multi-platform content publishing
🏢 Business and Sales • Customer support and CRM management • Automatic sales lead follow-up • Meeting scheduling and notes 🤖 Multi-Agent Collaboration • Agent team project management • Automated workflow orchestration • Shared knowledge bases and retrieval 🔧 More Scenarios • Smart home control • Financial data analysis • Education and training support

🙏 Contributors

Name Role Bio
Zixuan Gui Core Contributor -
Zhimin Zhao Core Contributor -
Xiuqi Li Core Contributor -
Lixin Liu Core Contributor -
Siyi Liu Core Contributor -
Sanbu Core Contributor -

More contributors are welcome

🤝 Contributing

  • If you find an issue, feel free to open an Issue. If you do not receive a response, you can contact the support team for follow-up.
  • If you would like to contribute, feel free to submit a Pull Request. If you do not receive a response, you can contact the support team for follow-up.
  • If you are interested in Datawhale and want to start a new project, please follow the Datawhale Open Source Project Guide.

💬 Community

Join the Hello Claw community group and learn together with other developers:

Community QR Code

📧 Follow Us

Scan the QR code below to follow the Datawhale official account on WeChat

📄 LICENSE

<img alt="Creative Commons License" style="border-width:0" src="https://img.s

Core symbols most depended-on inside this repo

require_config
called by 3
docs/en/university/email-assistant/examples/yesterday_mail_report.py
require_config
called by 3
docs/cn/university/email-assistant/examples/yesterday_mail_report.py
decode_mime_words
called by 2
docs/en/university/email-assistant/examples/yesterday_mail_report.py
clean_text
called by 2
docs/en/university/email-assistant/examples/yesterday_mail_report.py
format_imap_date
called by 2
docs/en/university/email-assistant/examples/yesterday_mail_report.py
decode_mime_words
called by 2
docs/cn/university/email-assistant/examples/yesterday_mail_report.py
clean_text
called by 2
docs/cn/university/email-assistant/examples/yesterday_mail_report.py
format_imap_date
called by 2
docs/cn/university/email-assistant/examples/yesterday_mail_report.py

Shape

Function 47

Languages

Python98%
TypeScript2%

Modules by API surface

docs/en/university/email-assistant/examples/yesterday_mail_report.py12 symbols
docs/cn/university/email-assistant/examples/yesterday_mail_report.py12 symbols
docs/en/university/email-assistant/examples/imap_connect_test.py11 symbols
docs/cn/university/email-assistant/examples/imap_connect_test.py11 symbols
scripts/reading-progress.js1 symbols

For agents

$ claude mcp add hello-claw \
  -- python -m otcore.mcp_server <graph>

⬇ download graph artifact