RRepoGEO

REPOGEO REPORT · LITE

Windy3f3f3f3f/claude-code-from-scratch

Default branch main · commit 5e674776 · scanned 5/18/2026, 5:57:17 PM

GitHub: 1,225 stars · 368 forks

Scan history for this repo

Score trend below includes all ready runs (older left, newer right; scroll horizontally if needed). The table is collapsed by default—expand for newest-first rows, 10 per page.

Score trend (left → right: older → newer)

2 ready scans. Expand the table below for newest-first rows (10 per page, paginated).

AI VISIBILITY SCORE
27 /100
Critical
Category recall
0 / 2
Not recommended in any query
Rule findings
2 pass · 0 warn · 0 fail
Objective metadata checks
AI knows your name
1 / 3
Direct prompts that named your repo
HOW TO READ THIS REPORT

Action plan is what to do next — copy-pasteable changes prioritized by impact. Category visibility is the real GEO test: when a user asks an AI a brand-free question that should surface Windy3f3f3f3f/claude-code-from-scratch, does the AI actually recommend you — or your competitors? Objective checks verify the metadata signals AI engines weight first. Self-mention check detects whether AI even knows you exist by name.

Action plan — copy-paste fixes

3 prioritized changes generated by gemini-2.5-flash. Mark items done after you ship the fix.

OVERALL DIRECTION
  • highreadme#1
    Reposition the README H1 to clarify it's a tutorial

    Why:

    CURRENT
    # Claude Code From Scratch
    COPY-PASTE FIX
    # Claude Code From Scratch: A Step-by-Step Tutorial to Build Your Own Coding Agent
  • highreadme#2
    Clarify the repository description to emphasize "tutorial" and "build"

    Why:

    CURRENT
    Build your own Claude Code from scratch. 🔍 Claude Code 开源了 50 万行代码，读不动？用 ~4000 行 TypeScript / Python 从零复现核心架构，11 章分步教程带你理解 coding agent 精髓
    COPY-PASTE FIX
    A step-by-step tutorial to build your own Claude-like coding agent from scratch. Learn the core architecture of advanced AI coding assistants by implementing a ~4000-line TypeScript/Python replica, bypassing Claude Code's 500k lines.
  • mediumreadme#3
    Strengthen the opening paragraph of the README to highlight the tutorial aspect

    Why:

    CURRENT
    Claude Code 开源了 50 万行 TypeScript。读不动？**
本项目用 **~4300 行代码**（TypeScript 和 Python 两个版本分别实现）复现了 Claude Code 的核心架构...
    COPY-PASTE FIX
    Claude Code 开源了 50 万行 TypeScript，难以理解？本项目提供一份**分步教程**，用 **~4300 行代码**（TypeScript 和 Python 两个版本分别实现）从零复现 Claude Code 的核心架构，帮助你快速理解并构建自己的高级 AI 编码 Agent。

Category GEO backends resolved for this scan: google/gemini-2.5-flash, deepseek/deepseek-v4-flash

Category visibility — the real GEO test

Brand-free queries asked to google/gemini-2.5-flash. Did AI recommend you, or someone else?

Same questions for every model — switch tabs to compare answers and rankings.

Recall
0 / 2
0% of queries surface Windy3f3f3f3f/claude-code-from-scratch
Avg rank
—
Lower is better. #1 = top recommendation.
Share of voice
0%
Of all named tools, what % are you?
Top rival
OpenAI GPT-4 / GPT-3.5 Turbo
Recommended in 2 of 2 queries
COMPETITOR LEADERBOARD
  1. OpenAI GPT-4 / GPT-3.5 Turbo · recommended 2×
  2. Anthropic Claude 3 · recommended 1×
  3. Google Gemini · recommended 1×
  4. Meta Llama 3 · recommended 1×
  5. Mistral AI · recommended 1×
  • CATEGORY QUERY
    What are the steps to create a functional AI programming agent from scratch?
    you: not recommended
    AI recommended (in order):
    1. OpenAI GPT-4 / GPT-3.5 Turbo
    2. Anthropic Claude 3
    3. Google Gemini
    4. Meta Llama 3
    5. Mistral AI
    6. LangChain (langchain-ai/langchain)
    7. LlamaIndex (run-llama/llama_index)
    8. OpenAI Function Calling / Tool Use
    9. Docker
    10. ESLint (eslint/eslint)
    11. Pylint (pylint-dev/pylint)
    12. Prettier (prettier/prettier)
    13. Black (psf/black)
    14. TypeScript (microsoft/TypeScript)
    15. MyPy (python/mypy)
    16. argparse
    17. commander.js (tj/commander.js)
    18. Streamlit (streamlit/streamlit)
    19. Gradio (gradio-app/gradio)
    20. React (facebook/react)
    21. Vue (vuejs/core)
    22. Angular (angular/angular)
    23. VS Code Extensions API
    24. Flask (pallets/flask)
    25. Django (django/django)
    26. FastAPI (tiangolo/fastapi)
    27. Express (expressjs/express)
    28. NestJS (nestjs/nest)
    29. openai Python library (openai/openai-python)
    30. anthropic Python library (anthropics/anthropic-sdk-python)
    31. GitPython (gitpython-developers/GitPython)
    32. AWS
    33. EC2
    34. Lambda
    35. Sagemaker
    36. Google Cloud
    37. Compute Engine
    38. Cloud Functions
    39. Vertex AI
    40. Microsoft Azure
    41. Azure Functions
    42. Azure Machine Learning
    43. Kubernetes (kubernetes/kubernetes)

    AI recommended 43 alternatives but never named Windy3f3f3f3f/claude-code-from-scratch. This is the gap to close.

    Show full AI answer
  • CATEGORY QUERY
    How to understand the core architecture of an advanced AI coding assistant?
    you: not recommended
    AI recommended (in order):
    1. OpenAI GPT-4 / GPT-3.5 Turbo
    2. Anthropic Claude 3 (Opus/Sonnet/Haiku)
    3. Google Gemini (Ultra/Pro)
    4. Meta Llama 3 (8B/70B)
    5. Pinecone
    6. Weaviate
    7. Qdrant
    8. LangChain
    9. LlamaIndex
    10. Tree-sitter
    11. Roslyn
    12. Clang
    13. Pyright
    14. ESLint
    15. Prettier
    16. Language Server Protocol (LSP)
    17. VS Code API
    18. IntelliJ Platform SDK
    19. VS Code Extensions
    20. JetBrains Plugins
    21. React
    22. Vue
    23. Svelte
    24. AutoGPT
    25. BabyAGI

    AI recommended 25 alternatives but never named Windy3f3f3f3f/claude-code-from-scratch. This is the gap to close.

    Show full AI answer

Objective checks

Rule-based audits of metadata signals AI engines weight most.

  • Metadata completeness
    pass

  • README presence
    pass

Self-mention check

Does AI even know your repo exists when asked about it directly?

  • Compared to common alternatives in this category, what is the core differentiator of Windy3f3f3f3f/claude-code-from-scratch?
    pass
    AI did not name Windy3f3f3f3f/claude-code-from-scratch — likely talking about a different project

    AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

  • If a team adopts Windy3f3f3f3f/claude-code-from-scratch in production, what risks or prerequisites should they evaluate first?
    pass
    AI named Windy3f3f3f3f/claude-code-from-scratch explicitly

    AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

  • In one sentence, what problem does the repo Windy3f3f3f3f/claude-code-from-scratch solve, and who is the primary audience?
    pass
    AI did not name Windy3f3f3f3f/claude-code-from-scratch — likely talking about a different project

    AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

Embed your GEO score

Drop this badge into the README of Windy3f3f3f3f/claude-code-from-scratch. It auto-updates whenever the report is rescanned and links back to the latest report — easy public proof that you care about AI discoverability.

RepoGEO badge previewLive preview
MARKDOWN (README)
[![RepoGEO](https://repogeo.com/badge/Windy3f3f3f3f/claude-code-from-scratch.svg)](https://repogeo.com/en/r/Windy3f3f3f3f/claude-code-from-scratch)
HTML
<a href="https://repogeo.com/en/r/Windy3f3f3f3f/claude-code-from-scratch"><img src="https://repogeo.com/badge/Windy3f3f3f3f/claude-code-from-scratch.svg" alt="RepoGEO" /></a>
Pro

Subscribe to Pro for deep diagnoses

Windy3f3f3f3f/claude-code-from-scratch — Lite scans stay free; this card itemizes Pro deep limits vs Lite.

  • Deep reports10 / month
  • Brand-free category queries5 vs 2 in Lite
  • Prioritized action items8 vs 3 in Lite