“coding”

想要文档优先的开发方法论时使用 — Stream Coding：先写清楚再写代码，80% 时间花在文档上

Evaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics.

Pure Claude code investigation.

Use when writing or refactoring code.

This skill embodies the principles of "Clean Code" by Robert C.

Apply maintainable, readable, and secure coding practices when implementing or reviewing software.

Use this skill when debugging failing unit tests and fixing code so that the test suite passes.

Applies Uncle Bob's Clean Code method-extraction principles to C# code.

Use when implementing, modifying, refactoring, or reviewing code and the agent must follow explicit coding standards for simplicity, readability…

Spawn a sub-agent to handle complex coding tasks in the background

Dual-perspective code investigation.

clean-code, the rules for readable code.

Install the Pi coding agent.

全面代码评审：code-reviewer + architect 双独立通道并行评审，严重度评级 + 架构状态（CLEAR/WATCH/BLOCK）+ 确定性合入门禁。用户说「review this code / code review / 帮我审代码」或合入 PR 前使用。

Applies the client-side subset of Galactic engineering rules.

Use when implementing, fixing, refactoring, testing, or deleting code (not just text files).

Use this skill to detect and clean up dead code in this repo, starting with the project’s TypeScript dead-code scripts and validating that removals do not…

Explain coding problems, code behavior, review findings, architecture, or proposed changes in plain language.

Guides agents using Pi coding-agent: install, authenticate, run, configure, automate, and extend the pi CLI.

Refines code through the three lenses in `code-refinements` with verified, behavior-preserving edits.

Provides object-oriented design conventions.

Baseline coding conventions for readability, naming, immutability, KISS, DRY, and code-quality review.

Creative coding: p5.js sketches and Manim math/algorithm animations for visual art, generative graphics, and educational videos.

Enforces internal React and TypeScript coding standards using avoid/prefer rules.