# shareAI-lab/learn-claude-code

Bash is all you need -  A nano claude code–like 「agent harness」, built from 0 to 1

Repository: https://github.com/shareAI-lab/learn-claude-code
Canonical: https://ross.abutalabs.com/products/learn-claude-code
Homepage: https://learn.shareai.run
Language: Python
License: MIT
License Family: permissive
Topics: agent, claude-code, teaching, agent-development, ai-agent, claude, educational, llm, python, tutorial
Last push: 2026-08-26T16:38:22+00:00

## Health v2 (maintenance only)
Score: 63/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 35, longevity 30
- inputs: {"age_days": 430, "days_push": 7, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 75378, forks 12165 (observed 2026-08-28T04:12:21.518689+00:00)

## What it is
An educational repository that teaches how to build a minimal Claude Code-like AI agent harness from scratch in Python. It explains that agent capability comes from the model while the harness is the surrounding infrastructure, walking through building that vehicle step by step.

## Use cases
- learn how coding agents like Claude Code work internally
- build a minimal AI agent harness from scratch
- understand the difference between model agency and agent infrastructure
- study agent loop and tool-calling implementation
- teach a course on LLM agent development
- use as a reference when building a custom coding agent

## When to choose
- you want to learn agent architecture by reading and building code
- you need a small, understandable reference implementation of an agent harness
- you are teaching or self-studying LLM agent development

## When to avoid
- you need a production-ready agent framework with full features
- you want a polished CLI coding assistant rather than educational code
- you need multi-model, enterprise-grade agent orchestration

## Facets
- artifact type: learning-resource
- maturity: active
- function: agent-framework, llm-inference, developer-tools
- domain: artificial-intelligence, large-language-models, tutorials, developer-tools
- platform: python, cross-platform
- tags: claude-code, agent-harness, educational, tutorial, from-scratch, teaching, education, ai-agents

## Member repositories
- shareAI-lab/learn-claude-code (main) score 63

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:21.518689+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 2026-08-29T16:14:42.561325+00:00, confidence not recorded.
  - readme: https://github.com/shareAI-lab/learn-claude-code (fetched 2026-08-28T04:12:21.518689+00:00, sha 96d1448497a7)
  - homepage: https://learn.shareai.run (fetched 2026-08-28T17:45:05.607397+00:00, sha 44136fa355b3)
- Data as of 2026-08-30T08:39:29.467469+00:00.