🔄 Agentic Methodology · Execution Discipline

Enforce red-green-refactor with hard gates

Failing test first, watch it fail, minimal code, watch it pass, commit — no code born before tests.

intermediate~30 minAgent Power UsersAI-Native EngineersTeams Shipping With Agents

Steps

1Write the failing test FIRST; run it and watch it fail for the right reason
2Forbid any production code until a test demonstrably fails
3Write the MINIMAL code to pass; resist gold-plating between greens
4Refactor only on green; keep the cycle under ten minutes
5Commit at every green — granular history is your undo log
6Delete any production code discovered written before its test

Common Pitfalls

▲Tests written after code confirming implementation details instead of behavior
▲'I'll add tests later' accumulating as invisible debt

Commands

Install with skills CLI

$ npx skills add aniruddhaadak80/skills --skill execution-discipline-tdd-red-green-refactor

Install globally

$ npx skills add aniruddhaadak80/skills --skill execution-discipline-tdd-red-green-refactor -g

Tags

#tdd#testing#discipline#agentic-methodology#execution-discipline

Install this skill

$ npx skills add aniruddhaadak80/skills --skill execution-discipline-tdd-red-green-refactor -a opencode

SKILL.md

# Enforce red-green-refactor with hard gates

> Failing test first, watch it fail, minimal code, watch it pass, commit — no code born before tests.

## When to Use This Skill

Failing test first, watch it fail, minimal code, watch it pass, commit — no code born before tests.

## Steps

1. Write the failing test FIRST; run it and watch it fail for the right reason
2. Forbid any production code until a test demonstrably fails
3. Write the MINIMAL code to pass; resist gold-plating between greens
4. Refactor only on green; keep the cycle under ten minutes
5. Commit at every green — granular history is your undo log
6. Delete any production code discovered written before its test

## Common Pitfalls

- Tests written after code confirming implementation details instead of behavior
- 'I'll add tests later' accumulating as invisible debt

View source on GitHub ↗

Fresh implementer per task, brief files as single source of truth, two-stage review, capped fix loops.

🔄 Agentic Methodology·~40m

Open →

Verify before claiming done — evidence over claimsfoundation

Fresh-command proof for every completion claim; 'it should work' is a confession.

🔄 Agentic Methodology·~20m