Get Started with Web3

An open-source, bilingual, AI-native Web3 curriculum for beginners, builders, and agents.

11 modules, 62 lessons, AI Tutor, badges, SEO prerendering, a versioned REST API, llms.txt, and a read-only MCP server.

Live Site  |  All Articles  |  API  |  llms.txt  |  中文说明  |  Twitter

Why This Exists

Web3 is hard to learn because beginner tutorials, protocol docs, security warnings, and builder examples are scattered everywhere. This project turns that mess into a structured learning path that can be read by humans and queried by AI agents.

If this repo helps you learn or teach Web3, please star the project so more learners can find it.

Who This Is For

Project Boundaries

This repository is a curriculum and AI-readable knowledge layer first. It will add focused labs for modern Web3 concepts, but those labs are educational examples, not a production dApp starter kit.

This project is not investment advice, token promotion, exchange onboarding, or a virtual-currency business service. Wallet, SIWE, x402, certificate, sponsor, and paid-tool materials must be clearly labeled as demos, metadata, drafts, or future plans until the relevant production systems are actually live and verified.

What Is Inside

Live Site

Use the hosted learning platform:

https://beihaili.github.io/Get-Started-with-Web3/

Public AI entrypoints:

• REST API documentation
• Production REST API
• Production OpenAPI 3.1 document
• llms.txt
• AI manifest
• AI content index

Course Map

Quick Start

git clone https://github.com/beihaili/Get-Started-with-Web3.git
cd Get-Started-with-Web3
npm install
npm run dev

The dev server starts at http://localhost:5173/Get-Started-with-Web3/.

Useful commands:

npm run build      # production build, OG images, sitemap, prerender
npm test           # Vitest suite
npm run lint       # ESLint
npm run ai:verify  # verify public AI entrypoints and x402 metadata

AI Agents and MCP

This repository exposes a machine-readable content layer for AI agents:

npm run ai:index   # regenerate ai/manifest.json, ai/content-index.json, ai/llms.txt
npm run ai:publish # copy AI artifacts into public/
npm run ai:verify  # verify public AI entrypoints and x402 metadata
npm run api:dev    # start the local read-only REST API on port 3001
npm run mcp:web3   # start the local read-only stdio MCP server

Example agent workflow:

npm run mcp:web3

For MCP clients that accept an mcpServers JSON block, copy this config and replace cwd with your local repository path:

{
  "mcpServers": {
    "get-started-with-web3": {
      "command": "npm",
      "args": ["run", "mcp:web3"],
      "cwd": "/absolute/path/to/Get-Started-with-Web3"
    }
  }
}

Then connect an MCP client and use:

• search_web3_content to find relevant lessons and glossary entries.
• read_web3_lesson to read a lesson with citations.
• get_learning_path to get role-based beginner, builder, researcher, or investor paths.
• lookup_web3_glossary to explain Web3 terms.
• compose_web3_context to build a bounded context pack with citations.
• list_monetizable_tools to inspect future x402-ready paid-tool metadata.

Local MCP tools are read-only. They do not enforce payment, sign transactions, or perform chain operations. x402 fields are reserved metadata for future hosted paid tools.

The REST API exposes the same generated content index over versioned, cacheable JSON endpoints with no authentication. It is live on a dedicated Vercel Functions host; bhbtc.xyz remains the website host and is not the API base URL. See the API documentation or the production OpenAPI document.

AI artifacts are maintained as a stable v1 public surface through artifactContract.version; see AI-native v1 stability and monetization decision.

Tech Stack

• Frontend: React 19 + Vite + Tailwind CSS
• State: Zustand
• Routing: React Router
• Content: Markdown in zh/ and en/
• Rendering: react-markdown + remark-gfm + rehype-raw
• AI Tutor: Gemini with user-provided API key
• Agent layer: versioned REST API, @modelcontextprotocol/sdk, generated JSON artifacts, llms.txt
• SEO: sitemap generation, OG image generation, prerendering
• Testing: Vitest + happy-dom + Testing Library
• CI/CD: GitHub Actions to GitHub Pages

Contributing

Good first contributions:

• Fix outdated or unclear lesson content.
• Proofread English translations.
• Add quiz questions or glossary entries.
• Improve accessibility, mobile UX, or tests.
• Add examples, diagrams, and case studies.

Start here:

• Contributing guide
• Contributor ladder
• Contributor spotlight template
• Contributor spotlights
• Good first issues catalog
• Open good first issues
• All issues

Support and Sponsorship

This project keeps learning content free. Support helps fund maintenance, translation, new lessons, and AI-native tooling.

• GitHub Sponsors: github.com/sponsors/beihaili
• Buy Me a Coffee: buymeacoffee.com/beihaili
• Sponsor kit: docs/strategy/2026-05-14-sponsor-kit.md
• Disclosure review: docs/strategy/2026-05-18-donation-affiliate-disclosure-review.md
• ETH / EVM: 0xb7Cd29C78B97cdf0A6e6FFC9268d43eDc3eCa649
• BTC: 1GihPqLkjJNK2fzCMSicc9mLapKeGy3Vv1
• Binance: Register with referral link (20% fee rebate)

Some outbound links may be affiliate or sponsor links and are disclosed where used.

Community

• Twitter/X: @bhbtc1337
• WeChat group: Apply via Google Form
• GitHub Issues: Ask questions and give feedback

Contributors

License

MIT License — free to use, modify, and distribute.

Star History