Can it help if the Rust compiler is already auto-vectorizing my code?

Yes, it features an Autovec Auditor that inspects compiler output to determine if the generated assembly is optimal or if manual intrinsics are required to unlock more performance.

How does it ensure the generated SIMD code is correct?

The skill includes a validation phase that uses property-based testing (proptest) to compare the outputs of the new SIMD implementation against the original scalar reference code.

What happens if my hardware doesn't support the target SIMD feature?

The skill implements a tiered fallback strategy, ensuring that the code always includes a safe scalar implementation for hardware that lacks specific SIMD capabilities.

Does this skill support both Intel and ARM processors?

Yes, SIMD Optimize detects the host architecture and generates specific implementations for x86 (SSE/AVX/AVX-512) and AArch64 (NEON/SVE) with appropriate runtime dispatch.

SIMD Optimize for Rust

Name: SIMD Optimize for Rust
Author: ahrav

byahrav

•

Developer Tools

Accelerates Rust code by implementing platform-specific SIMD vectorization with automated hardware detection and performance validation.

SIMD Optimize is a specialized Claude Code skill designed to boost the performance of data-parallel Rust applications. It automates the complex process of hardware ISA detection for x86-64 and AArch64 architectures, identifies vectorizable code patterns, and generates optimized SIMD intrinsics for ARM NEON/SVE and Intel/AMD SSE/AVX/AVX-512. By providing a structured workflow that includes correctness testing with proptest and performance benchmarking, it ensures that manual vectorization is both safe and measurably faster than standard compiler auto-vectorization.

Key Features

01Pattern analysis for maps, reductions, searches, and scans

02Correctness validation using property-based testing against scalar references

03Performance auditing to identify and fix failed compiler auto-vectorization

04Generation of platform-specific intrinsics including AVX-512 and SVE

052 GitHub stars

06Automated ISA feature detection for x86 and ARM architectures

Use Cases

01Adding cross-platform SIMD support for x86 and ARM in Rust crates

02Converting scalar search or transformation logic into high-throughput vector code

03Optimizing hot loops in high-performance data processing engines

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ahrav/scratch-scanner-rs simd-optimize

For use in Claude.ai and ChatGPT

Download Skill