tianji-qingtian

tianji-qingtian / AI-Native

Public

AI-Native 开发工作流方法论：4-Phase 框架 + 空项目/已有源码两种场景 + 反模式速查。让 Claude Code 等 AI agent 在共享契约下并行干活。

ai-agent ai-native anthropic claude claude-code
14
1
100% credibility
Found May 01, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
AI Summary

A guide and templates for a structured workflow that coordinates multiple AI agents under human-defined contracts to accelerate software development.

How It Works

1
🔍 Discover the guide

You stumble upon a friendly guide that teaches how AI helpers can team up to build projects faster and smarter.

2
📖 Read the big picture

You learn about a simple 4-step plan where humans set goals and rules, and AI handles the detailed work together.

3
🧠 Connect to your AI buddy

You do a quick one-time setup by placing the special instructions where your AI assistant can easily find and use them.

4
Choose your adventure
🆕
New project

Begin with a blank slate and share your exciting idea.

🔄
Existing project

Bring structure to your current work by scanning and planning.

5
💬 Chat with your AI

Tell your AI helper your project idea or goal, and it guides you through refining, rule-setting, building, and growing.

🚀 Projects come alive

Your software project builds quicker and stays organized, with AI doing the heavy lifting under your direction.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 14 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is AI-Native?

AI-Native is a methodology for AI-native development on GitHub, structuring workflows so AI agents like Claude Code work in parallel under shared contracts. It provides a 4-phase framework—intent refinement, contract skeletons, machine-readable contract networks, and intent-driven execution—for starting empty projects or auditing existing codebases. Developers get copy-paste starters, anti-pattern checklists, and Claude skills that guide the process via simple CLI prompts like "use AI-native way to start this project."

Why is it gaining traction?

Unlike vague AI-native software engineering tips, it delivers concrete 30-50% speedups for greenfield projects and 10-30% for legacy ones, with machine-readable contracts like tests and linters as the single source of truth. The hook is its Claude Code integration: install a skill once, then prompt in any repo for guided phases, avoiding AI hallucinations through human-led constraints. It's a practical bridge for ai-native companies building agent-driven apps, standing out in ai-native GitHub repos like ai-native panaversity.

Who should use this?

Tech leads and engineers at ai-native foundations or ai-native companies who rely on Claude Code as their primary CLI for coding. Ideal for kicking off agent-heavy projects from scratch or incrementally adding AI-native ERP features to brownfield code. Skip if you're just occasionally tweaking code with AI—costs outweigh benefits.

Verdict

With 14 stars and 1.0% credibility score, it's early-stage but docs are solid and starter templates work out-of-box; worth a spin for Claude users experimenting with ai-native meaning in software engineering. Test on a side project before committing—low risk, real methodology gains possible.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.