coding-workflow

Pre-implementation checklist, incremental development workflow, and code organization patterns. Use before starting implementation to ensure readiness and maintain quality throughout.

SKILL.md

GitHub

coding-workflow

Pre-implementation checklist and incremental development workflow.

how to use

  • /coding-workflow Apply this workflow to all implementation work in this conversation.

  • /coding-workflow <task> Run through the pre-implementation checklist for the given task.

when to apply

Reference these guidelines when:

  • starting a new feature or significant change
  • unsure about approach before writing code
  • working on unfamiliar codebase areas
  • making changes that touch multiple files
  • refactoring existing code

rule categories by priority

priority category impact
1 pre-implementation critical
2 incremental development critical
3 code organization high
4 change management medium

quick reference

1. pre-implementation (critical)

Before writing any code, verify:

  • understand the requirement (restate it in your own words)
  • read existing code in the affected area
  • identify all files that will change
  • check for existing patterns to follow
  • verify no duplicate implementation exists
  • understand the test strategy
  • identify potential breaking changes
  • check for related open PRs or in-progress work

Architecture check:

  • does this belong in an existing module or needs a new one?
  • what are the dependencies? Will this create circular dependencies?
  • is the change additive (safe) or modifying (riskier)?
  • what's the rollback plan if something goes wrong?

2. incremental development (critical)

  • make the smallest possible change that adds value
  • compile/lint after every meaningful edit
  • commit after each logical step (not at the end)
  • test each change before moving to the next
  • keep working code working; never leave things in a broken state

Progression:

  1. make it work (correct behavior)
  2. make it right (clean code, proper patterns)
  3. make it fast (optimize only if needed, with measurements)
  • resist the urge to refactor unrelated code while implementing
  • if you discover technical debt, note it; don't fix it in the same PR
  • prefer boring, proven solutions over clever ones

3. code organization (high)

  • files should have a single responsibility
  • group related code by feature, not by type (prefer user/ over controllers/)
  • keep files under 300 lines; split if larger
  • functions should do one thing; keep under 30 lines
  • extract shared logic only when used in 3+ places (Rule of Three)
  • name things for what they do, not how they work
  • put constants near usage, not in global files (unless truly global)

4. change management (medium)

  • one concern per PR; don't mix features with refactoring
  • write the PR description before or during implementation, not after
  • if a change grows beyond expected scope, stop and reassess
  • leave the codebase better than you found it, but within scope
  • when in doubt, ask before implementing