This has been superseded by rust-phase-skills

Rust Programming Skill

A comprehensive Claude Code skill for idiomatic Rust development. Covers ownership, traits, generics, async/await, error handling, serde, tracing, macros, design patterns, and production best practices. Current for Rust 2024 edition.

Installation

git clone https://github.com/badbeta/Rust_programming_skill.git ~/.claude/skills/rust-programming

The skill auto-activates for any .rs file via the globs: "*.rs" frontmatter.

Activation

The skill activates automatically when:

• Working with .rs files
• Writing or reviewing Rust code
• Planning Rust project architecture

Only the core SKILL.md get loaded into context at first. Other files are used as needed.

Better results by adding this to the first project or refactor prompt:

Do not use agents for planning or implementation. Do it yourself.

ALWAYS load the rust-programming skill and read both SKILL.md and architecture.md before starting to plan architecture or writing any code. No exceptions, so do it right now! The core skill contains essential rules, decision frameworks, and key patterns. Use architecture.md to decide on workspace layout, crate boundaries, trait-based DI, and growing architecture stages. Then consult supporting files:

• Use language-patterns.md for ownership patterns, Cow<T>, borrow splitting, entry API, let-else, if-let chains, closure capture semantics, and ? operator chains with context
• Prefer Result/Option with ? over unwrap/expect, iterator chains over index loops, borrowing over cloning, and typed errors over string errors
• Read async-concurrency.md when designing async systems, choosing channels, using tokio, rayon, or actors, and understanding Pin/Unpin
• Use type-system.md for trait patterns, GATs, const generics, type state, sealed traits, and lifetime patterns
• Use quick-reference.md for String, Vec, HashMap, Iterator, Option, Result, File/Path operations, and common trait implementations — choose the right method, don't reinvent it
• Follow error-handling.md for thiserror/anyhow/color-eyre decisions, multi-layer error translation, and error conversion chains
• Follow serde-serialization.md for derive attributes, enum representations, custom serialization, and format selection
• Follow documentation.md for rustdoc conventions, doc tests, intra-doc links, and standard sections (Examples, Errors, Panics, Safety)
• Use testing.md for unit, integration, mockall, insta, proptest, cargo-fuzz, and async test patterns
• Use web-apis.md for axum extractors, middleware, WebSocket, authentication, and database integration
• Use deployment.md for cargo profiles, Docker builds, CI/CD, cross-compilation, tracing subscribers, and metrics
• Every public type must derive Debug, every public enum should use #[non_exhaustive], every unsafe block needs a // SAFETY: comment