skills.rest
hw-native-sys
hw-native-sysOFFICIAL
Explore all skills by this author →
GitHub

feature-dev

Develops PyPTO IR features including operators, passes, and transformations.

101|83|Updated Dec 31, 2025
Search Tags:#transformation#compiler#operator#pypto#ir development#pass
One-click install
npx skills add https://github.com/hw-native-sys/pypto --skill feature-dev-hw-native-sys
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: feature-dev
Source: https://github.com/hw-native-sys/pypto/tree/main/.claude/skills/feature-dev
Command: npx skills add https://github.com/hw-native-sys/pypto --skill feature-dev-hw-native-sys

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and scripts (resource) components.

What problem does it solve?

This Skill streamlines the development of new Intermediate Representation (IR) features for the PyPTO framework, ensuring consistency and efficiency in adding operators, passes, and transformations.

Core Features & Use Cases

  • IR Feature Development: Facilitates the creation of new operators, optimization passes, and IR transformations.
  • Guided Workflow: Provides clear steps from documentation review to implementation and testing.
  • Use Case: A performance engineer needs to add a new fused operator to accelerate a specific AI workload. They use this Skill to follow the established pattern for operator registration, C++ implementation, Python binding, and testing.

Quick Start

Use the feature-dev skill to add a new tensor operator named 'my_op' following the provided C++ and Python patterns.

Frequently Asked Questions about feature-dev

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I add a new operator to PyPTO IR?▼

To add a new operator to PyPTO IR, follow the established pattern for operator registration, implement the logic in C++, add the necessary Python bindings, and integrate testing to ensure correctness.

What is the process for creating a new transformation pass in PyPTO?▼

Creating a transformation pass in PyPTO involves reading the framework documentation, implementing the pass logic using the IR builder, and integrating tests to validate the transformation behavior.

How do I implement C++ code and Python bindings for PyPTO IR features?▼

You implement the core IR feature logic in C++ and subsequently expose it to Python by adding bindings, following PyPTO's established patterns for seamless framework integration and testing.

Does PyPTO IR development require specific testing integration for new passes?▼

Yes, developing new passes and transformations requires integrating testing to validate the IR modifications, ensuring the new features adhere to PyPTO's established patterns and maintain system consistency.

When do I need to use the IR builder for PyPTO transformations?▼

You need to use the IR builder when developing new transformations and passes to construct and manipulate the Intermediate Representation according to PyPTO's established structural patterns.

Related Skills

Explore More Software Engineering Skills→
hw-native-syshw-native-sys
34

multi-repo-setup

Clones or updates external repos into a build directory for CI parity.

Official
Advanced
shadowinlifeshadowinlife
2

ask-q8-future-plan

Aggregate forecasts, IR notes, and strategy disclosures to validate future planning against execution.

Community
Advanced
rainoftimerainoftime
17

llvm-backend-generator

Generate LLVM Intermediate Representation and build compiler backends.

Community
Intermediate
mohitmishra786mohitmishra786
159

llvm

Generate, optimize, and inspect LLVM Intermediate Representation using Clang, opt, and llc.

Community
Intermediate