Co-Pilot / 辅助式
更新于 a month ago

tanstack-agent-skills

DDeckardGer
0.0k
deckardger/tanstack-agent-skills
82
Agent 评分

💡 摘要

此技能提供使用 TanStack 生态系统构建应用程序的最佳实践。

🎯 适合人群

React 开发者全栈开发者软件架构师技术负责人AI 开发者

🤖 AI 吐槽:看起来很能打,但别让配置把人劝退。

安全分析中风险

风险:Medium。建议检查:是否执行 shell/命令行指令;是否发起外网请求(SSRF/数据外发);文件读写范围与路径穿越风险。以最小权限运行,并在生产环境启用前审计代码与依赖。

TanStack Agent Skills

Comprehensive best practices for building applications with the TanStack ecosystem. These skills provide AI coding agents with structured guidelines for TanStack Query, TanStack Router, and TanStack Start.

Available Skills

tanstack-query

Best practices for data fetching, caching, mutations, and server state management with TanStack Query (React Query).

32+ rules across 8 categories:

  • Query Keys (CRITICAL) — Proper key structure and organization
  • Caching (CRITICAL) — staleTime, gcTime, and invalidation patterns
  • Mutations (HIGH) — Optimistic updates, error handling
  • Error Handling (HIGH) — Error boundaries, retry logic
  • Prefetching (MEDIUM) — Intent-based and route prefetching
  • Infinite Queries (MEDIUM) — Pagination patterns
  • SSR Integration (MEDIUM) — Hydration and dehydration
  • Performance (LOW) — Select transforms, memoization

tanstack-router

Best practices for type-safe routing, data loading, search params, and navigation with TanStack Router.

30+ rules across 8 categories:

  • Type Safety (CRITICAL) — Router registration, type narrowing
  • Route Organization (CRITICAL) — File-based routing, route trees
  • Data Loading (HIGH) — Loaders, Query integration
  • Search Params (HIGH) — Validation, type inheritance
  • Navigation (MEDIUM) — Link component, active states
  • Code Splitting (MEDIUM) — Lazy routes, critical path
  • Preloading (MEDIUM) — Intent-based preloading
  • Route Context (LOW) — Dependency injection

tanstack-start

Best practices for full-stack React applications with TanStack Start, including server functions, middleware, SSR, and authentication.

29+ rules across 8 categories:

  • Server Functions (CRITICAL) — createServerFn patterns
  • Security (CRITICAL) — Input validation, CSRF protection
  • Middleware (HIGH) — Request/function middleware
  • Authentication (HIGH) — Sessions, route protection
  • SSR (MEDIUM) — Hydration safety, streaming
  • Error Handling (MEDIUM) — Server errors, redirects
  • File Organization (LOW) — Code separation patterns
  • Deployment (LOW) — Environment config

tanstack-integration

Best practices for integrating TanStack Query with TanStack Router and TanStack Start together.

13 rules across 4 categories:

  • Setup (CRITICAL) — QueryClient context, provider wrapping
  • Data Flow (HIGH) — Loader + Query patterns
  • Caching (MEDIUM) — Single source of truth
  • SSR (LOW) — Dehydration/hydration

Installation

npx add-skill tanstack-agent-skills

Or manually add to your project's .cursor/ or Claude Code configuration.

Usage

Skills activate automatically when AI agents detect relevant contexts:

  • Building data-driven React applications
  • Setting up routing for React apps
  • Creating full-stack applications with SSR
  • Implementing authentication flows
  • Optimizing data fetching patterns

Skill Format

Each skill follows the Agent Skills standard:

skills/
├── tanstack-query/
│   ├── SKILL.md          # Main skill instructions
│   └── rules/            # Individual rule files
│       ├── qk-array-structure.md
│       ├── cache-stale-time.md
│       └── ...
├── tanstack-router/
│   ├── SKILL.md
│   └── rules/
├── tanstack-start/
│   ├── SKILL.md
│   └── rules/
└── tanstack-integration/
    ├── SKILL.md
    └── rules/

Rule File Structure

Each rule file contains:

  1. Priority — CRITICAL, HIGH, MEDIUM, or LOW
  2. Explanation — Why this pattern matters
  3. Bad Example — Anti-pattern to avoid with explanation
  4. Good Example — Recommended implementation
  5. Context — When to apply or skip the rule

Contributing

Contributions welcome! Please ensure:

  • Rules are practical and battle-tested
  • Examples are clear and runnable
  • Priority levels are appropriate
  • Context helps agents decide applicability

Resources

License

MIT

五维分析
清晰度9/10
创新性7/10
实用性9/10
完整性8/10
可维护性8/10
优缺点分析

优点

  • 全面的最佳实践
  • 多个 TanStack 工具的结构化指南
  • 支持多种应用类型

缺点

  • 可能需要熟悉 TanStack 工具
  • 复杂性可能会让初学者感到不知所措
  • 仅限于 TanStack 生态系统

相关技能

react-best-practices

A
toolCo-Pilot / 辅助式
86/ 100

“看起来很能打,但别让配置把人劝退。”

remotion

A
toolCo-Pilot / 辅助式
84/ 100

“用代码创建视频?希望你的代码和过渡一样流畅!”

newbie-next

A
toolCo-Pilot / 辅助式
82/ 100

“看起来很能打,但别让配置把人劝退。”

免责声明:本内容来源于 GitHub 开源项目,仅供展示和评分分析使用。

版权归原作者所有 DeckardGer.