coding

Routes software tasks to reference guides covering implementation, refactoring, review, debugging, and design standards.

2|Updated May 14, 2026
One-click install
npx skills add https://github.com/AnthonyPoschen/agent-skills --skill coding-anthonyposchen
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: coding
Source: https://github.com/AnthonyPoschen/agent-skills/tree/main/skills/coding
Command: npx skills add https://github.com/AnthonyPoschen/agent-skills --skill coding-anthonyposchen

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Software work spans many distinct activities—writing new code, refactoring, reviewing diffs, debugging failures, and designing APIs—and each demands different guardrails. This Skill acts as a single entrypoint that routes any programming task to the right workflow reference, so coding agents apply consistent standards without loading irrelevant guidance. ## Core Features & Use Cases - Task-Based Routing: Directs implementation, refactoring, review, debugging, design, file organization, verification, and assertion-strategy work to dedicated reference files loaded on demand. - Shared Coding Standards: Provides fallback rules for control flow, functions, abstractions, state management, magic values, and comments when project conventions do not decide. - Phased Workflows: Enforces ordered processes such as baseline-then-subtract-then-rewrite for refactors and reproduce-then-fix-then-verify for debugging, with explicit completion checks. - Use Case: When asked to fix a failing CI test, the agent loads the debugging reference, reproduces the failure through the real path, applies the smallest root-cause fix, and reruns the scenario before declaring completion. ## Quick Start Use the coding skill to review this pull request diff and report findings ordered by severity.

Frequently Asked Questions about coding

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

FAQPage Schema
How do I structure a code review with an AI coding agent?▼

Load the review workflow reference, which prioritizes findings that could cause bugs, regressions, security issues, or broken contracts. Findings are output ordered by severity with file and line evidence, followed by open questions and a brief summary.

How should an AI agent approach refactoring legacy code?▼

The refactoring workflow establishes a test baseline first, subtracts dead code and accidental complexity, then rewrites toward a stated target design in committed phases. Behavior is preserved unless the user explicitly requests changes, and unrelated legacy code stays untouched.

When should assertions be used instead of error returns?▼

Assertions check internal invariants and impossible states, while error returns handle expected runtime failures like invalid input or I/O problems. Correctness-critical invariants get always-on production assertions; expensive diagnostic checks stay as debug assertions.

Does this coding guidance work with languages other than Go?▼

Yes, the shared standards are language-agnostic, covering control flow, functions, abstractions, and comments. Language-specific guidance exists for Go package layout and Zig assertion patterns, and project tooling conventions always take priority.

When should mocks be avoided in test verification?▼

Mocks should only be used when the real dependency is unavailable, unsafe, prohibitively costly, or nondeterministic. For database, framework, or service behavior, prefer a real local dependency exercised through the normal application path with assertions on actual stored state.