Claude Skill

shareAI-lab/claw0

A structured learning resource to build claw-AI agents from scratch using OpenClaw. Follow step-by-step sections for hands-on AI agent development.

Overview

Stars3,053
Forks354
LanguagePython
Last pushed2026-06-30
Last synced2026-07-03
View on GitHub

Repository

OwnershareAI-lab
Repositoryclaw0
Full nameshareAI-lab/claw0
Repo ID1,165,581,971

Install this Skill

git clone https://github.com/shareAI-lab/claw0.git && cd claw0

Registry

Typeopenclaw_skill
Quality score75/100
Verificationreadme_parsed
Last verified2026-05-31
Platforms
ClaudeOpenClaw
Capabilities
pdfmemory
Detected files
README.mdrequirements.txt
Config keys
ANTHROPIC_API_KEYMODEL_ID
Install methods
  • git clone https://github.com/shareAI-lab/claw0.git && cd claw0
  • pip install -r requirements.txt

Summary

Claw0 is a comprehensive learning resource designed to guide users from zero to building a claw-AI agent from scratch using OpenClaw. It provides structured sections and tutorials for hands-on development.

Chinese description

0 - 1 学习OpenClaw:从零构建爪式AI智能体的章节

Key features

  • Step-by-step learning from zero to one
  • Structured sections for building AI agents
  • Focus on claw-AI agent development
  • Hands-on, practical tutorials
  • Uses the OpenClaw framework

Use cases

  • Learning AI agent development from scratch
  • Building specialized claw-AI agents
  • Educational resource for AI and robotics
  • Practical project-based learning
  • Experimenting with OpenClaw framework

README excerpt

[English](README.md) | [中文](README.zh.md) | [日本語](README.ja.md) # claw0 **From Zero to One: Build an AI Agent Gateway** > 10 progressive sections -- every section is a single, runnable Python file. > 3 languages (English, Chinese, Japanese) -- code + docs co-located. --- ## What is this? Most agent tutorials stop at "call an API once." This repository starts from that while loop and takes you all the way to a production-grade gateway. Build a minimal AI agent gateway from scratch, section by section. 10 sections, 10 core concepts, ~7,000 lines of Python. Each section introduces exactly one new idea while keeping all prior code intact. After all 10, you can read OpenClaw's production codebase with confidence. ```sh s01: Agent Loop -- The foundation: while + stop_reason s02: Tool Use -- Let the model call tools: dispatch table s03: Sessions & Context -- Persist conversations, handle overflow s04: Channels -- Telegram + Feishu: real channel pipelines s05: Gateway & Routing -- 5-tier binding, session isolation s06: Intelligence -- Soul, memory, skills, prompt assembly s07: Heartbeat & Cron -- Proactive agent + scheduled tasks s08: Delivery -- Reliable message queue with backoff s09: Resilience -- 3-layer retry onion + auth profile rotation s10: Concurrency -- Named lanes serialize the chaos ``` ## Architecture ``` +------------------- claw0 layers -------------------+ | | | s10: Concurrency (named lanes, generation track) | | s09: Resilience (auth rotation, overflow compact)| | s08: Delivery (write-ahead queue, backoff) | | s07: Heartbeat (lane lock, cron scheduler) | | s06: Intelligence (8-layer prompt, h

Topics

No topics yet.

Explore more

Data from GitHub. Synced on 2026-07-03