# datawhalechina/hello-agents

📚 《从零开始构建智能体》——从零开始的智能体原理与实践教程

Repository: https://github.com/datawhalechina/hello-agents
Canonical: https://ross.abutalabs.com/products/hello-agents
Homepage: https://hello-agents.datawhale.cc
Language: Python
License: NOASSERTION
License Family: other
Topics: agent, tutorial, llm, rag
Last push: 2026-08-18T06:51:35+00:00

## Health v2 (maintenance only)
Score: 73/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 98, release rhythm 69, longevity 25
- inputs: {"age_days": 360, "days_push": 15, "days_rel": 47, "gap_med": 127.5, "n_releases_24m": 3}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 75072, forks 9325 (observed 2026-08-28T04:12:21.500196+00:00)

## What it is
Hello-Agents is a free, open-source tutorial from the Datawhale community that teaches how to build AI-native intelligent agents from scratch, covering core principles, classic paradigms, and hands-on implementation. It includes building a custom agent framework on OpenAI-native APIs, plus advanced topics like memory, context engineering, evaluation, and agentic RL training.

## Use cases
- learn how to build AI agents from scratch
- understand agent architectures and paradigms
- build a multi-agent application hands-on
- train LLMs with agentic RL from SFT to GRPO
- prepare for AI agent job interviews
- learn RAG and memory systems for agents
- study context engineering and agent evaluation

## When to choose
- you want a structured, theory-plus-practice agent course
- you prefer learning by building your own framework rather than using low-code tools
- you want free community-maintained educational content on LLM agents

## When to avoid
- you need a production-ready agent framework rather than a tutorial
- you only want drag-and-drop low-code agent builders like Dify or Coze
- you need a permissively licensed library for commercial use

## Facets
- artifact type: learning-resource
- maturity: active
- function: agent-framework, rag, llm-training, prompt-engineering, developer-tools
- domain: large-language-models, tutorials, education
- platform: python, cross-platform
- tags: tutorial, agents, llm, datawhale, open-course, chinese, hands-on, multi-agent, ai-agents, retrieval-augmented-generation

## Member repositories
- datawhalechina/hello-agents (main) score 73

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:21.500196+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:14:42.861324+00:00, confidence not recorded.
  - readme: https://github.com/datawhalechina/hello-agents (fetched 2026-08-28T04:12:21.500196+00:00, sha c65bac04da89)
  - homepage: https://hello-agents.datawhale.cc (fetched 2026-08-28T17:45:25.929937+00:00, sha 59a814988e26)
- Data as of 2026-08-30T08:39:29.467469+00:00.