Skip to content

snowdreamtech/template

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1,489 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Snowdream Tech AI IDE Template

GitHub Actions Lint GitHub Actions Verify GitHub Release License: MIT CodeSize Dependabot Enabled

English | 简体中文

An enterprise-grade, foundational template designed for multi-AI IDE collaboration. This repository serves as a Single Source of Truth for AI agent rules, workflows, and project configurations, supporting over 50 different AI-assisted IDEs with massive multi-language support.

🌟 Features

  • Multi-IDE Compatibility: Out-of-the-box support for Cursor, Windsurf, GitHub Copilot, Cline, Roo Code, Trae, Gemini, Claude Code, and 50+ other AI editors.
  • Unified Rule System: Centralized rule definitions in .agent/rules/. Changes propagate automatically to all supported IDEs via a secure symlink/redirect pattern.
  • 80+ Language & Framework Rules: Pre-configured high-quality rules from Rust, Go, TypeScript, Python to Ansible, Kubernetes, and API design.
  • Smart Workflows (SpecKit): Standardized .agent/workflows/ commands (speckit.plan, speckit.analyze, snowdreamtech.init) that behave consistently across all supported environments.
  • Triple Guarantee Quality: 100% code purity enforced through Pre-commit and GitHub Actions integrated quality gates.
  • Cross-Platform Ready: Runs seamlessly on macOS (Homebrew/MacPorts), Linux, and Windows.

🏗️ Section 1 — Design & Architecture

Overview

The Snowdream Tech Template is a foundational scaffold engineered to solve the "N-IDE Fragmentation" problem. It standardizes the development environment, AI agent rules, and automation pipelines across varied platforms and languages.

Key Capabilities:

  • Provides a Unified Rule Engine that governs AI behavior consistently across 50+ IDEs.
  • Enforces Cross-Platform Portability through dynamically adapting POSIX shell automation.
  • Implements a Triple Guarantee Quality Gate (IDE, CLI, CI) to prevent regressions.
  • Supports Massive Multi-Language Stacks with modular onboarding logic.

Architecture

graph TD
    A["Developers & Agents"] -->|Operates via| IDE["Cursor / Windsurf / Copilot / 50+ Others"]
    IDE -->|Reads Rules via Redirects| R1[".vscode/"]
    IDE -->|Reads Rules via Redirects| R2[".github/"]
    IDE -->|Reads Rules via Redirects| R3[".cline/ .trae/ etc."]

    R1 -.->|SSoT Pointer| CoreRules[".agent/rules/"]
    R2 -.->|SSoT Pointer| CoreRules
    R3 -.->|SSoT Pointer| CoreRules

    CoreRules -->|Governs| Src["Source Code"]
    CoreRules -->|Governs| Scripts["CI/CD & Shell Scripts"]
Loading

Design Principles

  • Single Source of Truth (SSoT): All AI rules, commands, and Git hooks live in one place. No duplicated IDE configurations.
  • Cross-Platform Portability: Heavy automation logic is written in POSIX Shell, with thin wrappers for Windows PowerShell/Batch.
  • Triple Guarantee Quality: Linting and formatting form an impenetrable wall, enforced at the IDE layer, pre-commit layer, and CI/CD GitHub Actions layer.

Responsibilities

  • .agent/rules/: Owns the definitive behavioral logic for AI agents across all supported languages.
  • scripts/: Owns the cross-platform automation and lifecycle logic.
  • .agent/workflows/: Owns the interactive AI commands (SpecKit).

📖 Section 2 — Usage Guide

Prerequisites

  • Runtime: Node.js (>= 20.x), Python (>= 3.10.x).
  • Git: Global git installation required.

Quick Start

  1. Prerequisites: mise is highly recommended for global tool management (automatically installed during setup).
  2. Initialize: make setup (bootstraps mise and core tools).
  3. Install: make install (installs project dependencies).
  4. Verify: make verify (ensures everything is green).

Configuration Reference

Parameter Purpose Location
PROJECT_NAME Project identity init-project.sh
GITHUB_PROXY Network optimization (See Proxy Usage) scripts/lib/common.sh
VERSION Semantic versioning package.json

File Structure

project-root/
├── .agent/              # 🤖 Canonical AI configuration (The Brain)
│   ├── rules/           # 📏 Unified AI behavioral rules (80+ sets, SSoT)
│   └── workflows/       # 🛠️ Unified commands & AI workflows (SpecKit)
├── .agents/             # 🧩 Shared command sources (Auto-managed symlinks)
├── .github/             # 🐙 GitHub integration & Copilot settings
├── .vscode/             # 💻 Optimized VS Code configurations
└── src/                 # 📦 Your actual application source code

🛠️ Section 3 — Operations Guide

Pre-deployment Checklist

  1. Run make verify to ensure all quality gates are green.
  2. Run make audit to verify security compliance.
  3. Ensure CHANGELOG.md is updated.

Performance Considerations

  • Linting Speed: Pre-commit hooks target < 5s by scanning staged files only.
  • CI Throughput: GitHub Actions use matrix builds for parallel testing across OS types.

Troubleshooting

  • Problem: make install fails on Windows.
    • Diagnosis: Check if ExecutionPolicy allows script execution.
    • Solution: Run Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass.
  • Problem: Gitleaks detects false positives.
    • Diagnosis: Check .gitleaks.toml allowlist.
    • Solution: Add fingerprint to .gitleaksignore.
  • Problem: Pre-commit hooks fail on macOS after make install with Python errors.
    • Diagnosis: Check if the venv exists: ls .venv/bin/python.
    • Solution: Rebuild the venv: rm -rf .venv && make install.

🔒 Section 4 — Security Considerations

Security Model

  • Secret Management: All secrets must be injected via environment variables or handled by HashiCorp Vault. Never commit .env files.
  • Audit Logging: All critical operations (commits, releases, state changes) are traced via Git and CI logs.
  • Supply Chain: All CI actions are pinned to exact versions/SHAs.

Best Practices

Aspect Requirement Implementation
Secrets No plaintext secrets in repo gitleaks enforced at commit
Integrity Verify downloads SHA-256 validation in common.sh
Permissions Non-root execution Dockerfile best practices

🧑‍💻 Section 5 — Development Guide

Code Organization

project-root/
├── .agent/               # AI configuration (Single Source of Truth)
│   ├── rules/            # 88 behavioral rule files for AI agents
│   └── workflows/        # SpecKit slash-command definitions
├── .github/              # GitHub ecosystem (Actions, templates, Dependabot)
│   └── workflows/        # CI/CD pipelines (lint, verify, release, security)
├── .devcontainer/        # DevContainer configuration for reproducible environments
├── docs/                 # Project documentation
│   ├── adr/              # Architecture Decision Records
│   ├── runbooks/         # Operations and recovery runbooks
│   └── glossary.md       # Bilingual term glossary
├── scripts/              # POSIX shell automation (setup, install, verify)
│   └── lib/              # Shared shell library functions
└── Makefile              # Task orchestration (setup, install, lint, verify, audit)

Naming Conventions: Rule files use NN-kebab-case.md (core rules) or technology.md (language stacks). Workflow files use namespace.verb.md. Shell scripts use kebab-case.sh.

Extension Points

  1. Adding Rules: Create a new .md file in .agent/rules/ and link it in 00-index.md.
  2. Adding Commands: Add .md files to .agent/workflows/.
  3. Adding IDE Support: Create a redirect folder (e.g., .myide/) following the symlink pattern in Rule 03.

Local Development Setup

git clone <repo>
cd <repo>
git config core.ignorecase false  # MANDATORY for Mac/Windows
make setup
make install

References

🚀 Proxy Usage Scenarios

The GITHUB_PROXY (default: https://gh-proxy.sn0wdr1am.com/) is optimized for specific network acceleration scenarios. Misusing it for unsupported protocols (like Git) will result in errors.

Scenario Supported? Example / Note
Release Files ✅ Yes .../releases/download/v1.0/tool.zip
Source Archives ✅ Yes .../archive/master.zip or .tar.gz
Direct File Links ✅ Yes .../blob/master/filename
Git Clone No Do not use for git clone or insteadOf configs.
Project Folders No Browsing/cloning via proxy is not supported.

Important

To prevent breaking toolchains (like mise or asdf), this template explicitly disables Git redirection via this proxy. Use it only for direct HTTP downloads in scripts.

📄 License

This project is licensed under the MIT License. Copyright (c) 2026-present SnowdreamTech Inc. See the LICENSE file for the full license text.

About

An enterprise-grade, foundational template designed for multi-AI IDE collaboration. This repository serves as a Single Source of Truth for AI agent rules, workflows, and project configurations, supporting over 50 different AI-assisted IDEs with massive multi-language support.

Topics

Resources

License

Code of conduct

Contributing

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors