s-hiraoku/codex-harnesses · Archived

tdd

Drive implementation through red-green-refactor with isolated test design, minimal implementation, and behavior-preserving cleanup.

First seen Jun 13, 2026

Installation

$ npx skills add s-hiraoku/codex-harnesses --skill tdd

Stronger alternatives

This repository is archived — consider an actively maintained alternative.

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from s-hiraoku/codex-harnesses · top by installs.

npx skills add s-hiraoku/codex-harnesses

Browse all from s-hiraoku/codex-harnesses

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Repository health

License LICENSE
Default branch main
Open issues 2
Status Archived

Skill metadata

Parsed from SKILL.md frontmatter.

Declared agents codex

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 1,648 B
  • docs SUMMARY.md 142 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 2 installs

SKILL.md

Test-Driven Development

Use this workflow when the user wants new behavior developed under TDD discipline, or when a risky behavior change benefits from a failing test before implementation.

Workflow

  1. Specify the behavior in one or two concrete sentences and identify the test framework.
  2. Red: write one focused test that captures the requested behavior. Prefer existing test patterns and fixtures. Run it and confirm it fails for the expected reason.
  3. Green: implement the smallest code change that makes the failing test pass. Avoid speculative options, broad rewrites, or unrelated cleanup.
  4. Run the focused test. If it still fails, fix the implementation rather than weakening the test unless the test contradicts the stated behavior.
  5. Refactor: clean only the code introduced or directly touched by the change. Preserve public APIs, return values, side effects, and data formats.
  6. Run the focused test again, then related tests for the touched area.
  7. Repeat the loop for each distinct behavior instead of writing a broad test batch up front.

When subagents are available, keep phases isolated: pass the behavior only to a test-writing subagent, pass the failing test only to an implementation subagent, and pass the green code only to a refactoring subagent.

Final Report

Include:

  • behavior implemented
  • red test added or updated
  • implementation files changed
  • refactor changes applied
  • tests run and result
  • remaining risks or follow-up tests