AgentSkillsCN

rust-core

全面掌握 Rust 开发技能,涵盖核心原理、设计模式、错误处理、异步编程、测试与性能优化。适用于以下场景:(1) 需要指导的 Rust 项目,例如:(2) 掌握语言基础(所有权、生命周期、借用机制);(3) 做出架构决策与设计模式选择;(4) 开发 Web 应用(Axum、Actix-web、Rocket);(5) 集成 AI/大语言模型;(6) 构建 CLI/TUI 应用程序;(7) 使用 Tauri 开发桌面应用;(8) 掌握 async/await 与并发编程;(9) 制定错误处理策略;(10) 进行测试与基准测试;(11) 优化性能;(12) 实施日志记录与可观测性;(13) 开展代码审查并践行最佳实践。

SKILL.md
--- frontmatter
name: rust-core
description: "Comprehensive Rust development expertise covering core principles, patterns, error handling, async programming, testing, and performance optimization. Use when working on Rust projects requiring guidance on: (1) Language fundamentals (ownership, lifetimes, borrowing), (2) Architectural decisions and design patterns, (3) Web development (Axum, Actix-web, Rocket), (4) AI/LLM integration, (5) CLI/TUI applications, (6) Desktop development with Tauri, (7) Async/await and concurrency, (8) Error handling strategies, (9) Testing and benchmarking, (10) Performance optimization, (11) Logging and observability, or (12) Code reviews and best practices."

Rust Core Development

Comprehensive guidance for Rust development across web, CLI, desktop, AI/LLM applications, and systems programming.

Quick Reference Guide

By Task Type

Getting Started

  • New Project Setup: Use scripts/init_rust_project.sh to scaffold a project with best practices
  • Core Principles: See references/principles.md for ownership, borrowing, and Rust fundamentals
  • Common Errors: See references/common-errors.md for solutions to frequent compiler errors

Writing Code

Domain-Specific Development

Code Quality

Project Management

By Question Type

QuestionReference
"How do I handle errors?"error-handling.md
"Which web framework should I use?"web-frameworks.md
"How do I work with async/await?"async-patterns.md
"How do I integrate OpenAI/Claude?"ai-llm.md
"How do I build a CLI?"cli-tui.md
"How do I create a desktop app?"desktop-tauri.md
"Why won't this compile?"common-errors.md
"How do I improve performance?"performance.md
"How do I add logging?"logging-observability.md
"How should I name this?"naming.md
"What are best practices?"principles.md
"How do I test this?"testing.md

Core Workflows

1. Starting a New Project

  1. Initialize Project

    bash
    ./scripts/init_rust_project.sh my-project
    
  2. Set Up Development Tools

    • Configure linting: Copy assets/configs/clippy.toml and assets/configs/rustfmt.toml
    • Configure security: Copy assets/configs/deny.toml
    • Run audit: ./scripts/audit_dependencies.sh
  3. Add Logging

    bash
    ./scripts/setup_logging.sh
    
  4. Choose Architecture

2. Implementing Features

  1. Design First

  2. Write Code

    • Follow naming conventions from naming.md
    • Use appropriate patterns and error handling
    • Add tracing/logging as you go
  3. Test

    • Write unit tests (see testing.md)
    • Add integration tests for public APIs
    • Consider property-based tests for complex logic

3. Code Review and Refinement

  1. Self-Review

    • Run through code-review.md checklist
    • Check for common anti-patterns
    • Verify error handling
  2. Performance Check

    • Profile if performance-critical (see performance.md)
    • Benchmark changes with Criterion
    • Avoid premature optimization
  3. Security Audit

    bash
    ./scripts/audit_dependencies.sh
    

Decision Guides

Choosing a Web Framework

Use Axum when:

  • Building modern REST/GraphQL APIs
  • Want composable middleware (Tower ecosystem)
  • Prefer type-driven extractors
  • Building microservices

Use Actix-web when:

  • Need maximum performance
  • Building high-throughput APIs
  • Want mature, battle-tested framework
  • Familiar with actor model

Use Rocket when:

  • Rapid prototyping
  • Want batteries-included features
  • Smaller team or learning Rust web
  • Traditional web application

See web-frameworks.md for detailed comparison and code examples.

Error Handling Strategy

Use anyhow for:

  • Applications (binaries)
  • Quick prototyping
  • Internal tools
  • When you need ergonomic error handling with context

Use thiserror for:

  • Libraries (public APIs)
  • When consumers need to handle specific error cases
  • Type-safe error hierarchies
  • Production code with well-defined error types

See error-handling.md for patterns and examples.

When to Use Async

Use async/await when:

  • I/O-bound operations (network, file system)
  • Web servers handling many concurrent requests
  • Database connection pooling
  • Working with streams of data

Don't use async when:

  • CPU-bound operations (use spawn_blocking instead)
  • Simple CLI tools
  • Performance isn't critical
  • Complexity isn't justified

See async-patterns.md for Tokio patterns and best practices.

Automation Scripts

Available Scripts

scripts/init_rust_project.sh Initialize a new Rust project with best practices:

  • Common dependencies (anyhow, thiserror, serde, tracing)
  • Benchmark setup with Criterion
  • Optimized release profile
  • Proper .gitignore

Usage: ./scripts/init_rust_project.sh my-project [bin|lib]

scripts/audit_dependencies.sh Audit dependencies for security and licensing:

  • Runs cargo-audit for security vulnerabilities
  • Runs cargo-deny for license compliance
  • Shows outdated dependencies

Usage: ./scripts/audit_dependencies.sh

scripts/setup_logging.sh Set up tracing-based logging:

  • Adds tracing dependencies
  • Creates logging module with JSON support
  • Provides initialization code

Usage: ./scripts/setup_logging.sh

Configuration Templates

assets/configs/clippy.toml

Clippy linting configuration for strict code quality

assets/configs/rustfmt.toml

Code formatting configuration (100 char width, Unix newlines)

assets/configs/deny.toml

cargo-deny configuration for:

  • Security advisory checking
  • License compliance (MIT, Apache-2.0, BSD allowed)
  • Duplicate dependency detection
  • Source verification

Reference Documentation

All reference files provide in-depth guidance on specific topics:

Core Language

Development

Async & Concurrency

Domains

Libraries

  • crates-core.md - Essential crates (serde, tokio, anyhow, reqwest)

When to Consult References

Load references progressively as needed:

  1. Starting out: Read principles.md to understand Rust fundamentals
  2. Choosing approach: Consult domain-specific guides (web-frameworks.md, ai-llm.md, etc.)
  3. Implementing: Reference patterns.md and error-handling.md
  4. Debugging: Check common-errors.md
  5. Optimizing: See performance.md
  6. Reviewing: Use code-review.md checklist

Don't load all references at once—consult them as specific needs arise.

Best Practices Summary

  1. Embrace the borrow checker - Work with ownership, not against it
  2. Use type-driven design - Make invalid states unrepresentable
  3. Handle errors explicitly - Use Result and Option, avoid unwrap()
  4. Test comprehensively - Unit tests, integration tests, property tests
  5. Profile before optimizing - Measure, don't guess
  6. Add structured logging - Use tracing for observability
  7. Review security - Audit dependencies, validate inputs
  8. Follow conventions - rustfmt, clippy, naming conventions
  9. Document public APIs - Doc comments with examples
  10. Keep it simple - Prefer clarity over cleverness