CodeWhale was formerly known as DeepSeek-TUI.
Learn CodeWhale through a practical 5-stage learning path built for beginners who want a hands-on CodeWhale Tutorial. This course helps you understand DeepSeek TUI concepts, daily AI coding workflows, and the CodeWhale CLI from the ground up.
CodeWhale is an AI coding command-line environment for learning, building, and shipping with agentic workflows. This repo turns the original zero2deepseek-tui skillpack into a public learning path for people searching for "Learn CodeWhale", "CodeWhale Tutorial", "DeepSeek TUI", and "CodeWhale CLI".
Use it when you want to:
- Learn CodeWhale step by step instead of reading scattered notes.
- Practice core CLI workflows with guided tasks.
- Understand modes, configuration, skills, MCP, memory, thinking, and subagents.
- Build a small CLI tool as a capstone project.
| Stage | Folder | Focus |
|---|---|---|
| 1 | stages/stage-01-meet-codewhale |
Meet CodeWhale and understand the basic workflow. |
| 2 | stages/stage-02-modes-and-config |
Learn modes, configuration, and practical setup habits. |
| 3 | stages/stage-03-skills-mcp-memory |
Use skills, MCP integrations, and memory effectively. |
| 4 | stages/stage-04-thinking-and-subagents |
Apply thinking workflows and subagent patterns. |
| 5 | stages/stage-05-build-cli-tool |
Build a CLI tool to consolidate what you learned. |
Each stage includes a README.md for concepts and a tasks.md file for hands-on practice.
- Basic command-line experience.
- A local development environment with Git installed.
- Interest in AI coding workflows and CLI-based learning.
- No prior CodeWhale or DeepSeek TUI experience required.
git clone https://github.com/aiflowlearn/zero2codewhale.git
cd zero2codewhaleThen start with Stage 1:
open stages/stage-01-meet-codewhale/README.md
open stages/stage-01-meet-codewhale/tasks.mdRecommended flow:
- Read the stage concept
README.md. - Complete the matching
tasks.mdexercises. - Review related notes in
knowledge/. - Move to the next stage only after you can explain the workflow in your own words.
You can practice this course in any environment where you can read markdown files and use a terminal. For best results:
- Keep a practice repository for your CodeWhale experiments.
- Record the prompts, commands, and decisions you make.
- Repeat each stage with a small personal project.
- Use the knowledge docs as quick references while working.
CodeWhale was formerly known as DeepSeek-TUI. This repo preserves the historical learning material while using the current CodeWhale name.
Yes. It starts from basic concepts and gradually introduces more advanced workflows such as skills, MCP, memory, thinking, and subagents.
For the best learning outcome, yes. The stages are ordered so that later tasks build on earlier habits.
No. The course materials are licensed under CC BY-NC-SA 4.0, which allows sharing and adaptation for non-commercial use only.
- awesome-claude-code — Awesome Claude Code resources
- zero2codewhale — Zero to Hero: Learn CodeWhale from scratch
- 9router-starter-kit — 9router starter kit
- ai-coding-projects — AI coding projects
- ai-agent-skillpack — AI agent skillpack
- geo-llm-brand-playbook — GEO LLM brand playbook
- codewhale-examples — CodeWhale examples
This repository is licensed under CC BY-NC-SA 4.0 only. It contains learning materials and does not include tools or scaffolding under a separate software license.
Sponsored by AIFlowLearn — AI-native learning platform
AIFlowLearn / AI智流学社 — Browser-based AI coding practice environments, interactive courses, and skill packs.