KairaxisKairaxis
Freestructured workflowlow risk

Capability · rank 12

tdd

Drive implementation through red-green-refactor with practical test-first habits.

Guides an agent through red-green-refactor loops and practical test-first development.

Listed owner

mattpocock/skills

Package source

GitHub

Category

Developer Exp

Package validatedBenchmark testedCreator reviewedDeveloper Exp
$npx skills add mattpocock/skills --skill tdd
OverviewSandboxUsageExamplesLimitationsVersionsRelated Skills

Overview

Guides an agent through red-green-refactor loops and practical test-first development. Review the sandbox comparison, limitations, permissions, and pinned version before install.

Outcome

Drive implementation through red-green-refactor with practical test-first habits.

Who it is for

Agents implementing features that need durable tests.

What it is not for

Exploratory spikes where discarding code is expected.

Runtime export

codex_skill · markdown_export

Sandbox comparison
Generic baseline vs tdd on the same sample input · pinned v2.0.0

Sample input

Add price formatting helper used in checkout and invoices.

Model: gpt-4.1-mini · temp 0.2

Generic baseline

48/100
Write the helper, then add a couple of tests if time allows.

Skill-enhanced

94/100
Red: failing cases for zero, negative, major/minor units, and locale.
Green: minimal formatter.
Refactor: extract currency strategy and freeze fixtures.
Do not expand scope into tax calculation.

Rubric scores

+46 lift
Coverage of failure modes2 → 5
Actionable recommendations3 → 5
Measurable next steps2 → 4
Risk / limitation awareness2 → 4

Assumptions

  • Sample input is representative of a real production task
  • Same model and temperature for baseline and skill-enhanced runs

Comparison runs through the shared sandbox engine on registry sample cases (deterministic v0 path).

Usage

  1. 1. Inspect trust badges, limitations, and permissions on this page.
  2. 2. Run the sandbox comparison to see baseline vs skill-enhanced output.
  3. 3. Install the pinned version with the command below for your agent runtime.
  4. 4. Agents can also fetch /api/v1/skills/tdd/manifest.
$npx skills add mattpocock/skills --skill tdd

Example prompts & use cases

1 / Prompt

Implement this function with TDD

2 / Prompt

Convert this bug fix into a failing test first

Limitations

  • Cannot invent business requirements missing from the brief.

Audience

Agents implementing features that need durable tests.

Not for

Exploratory spikes where discarding code is expected.

Version history

v2.0.0

Jan 9, 2026current

Stronger red-green-refactor gating

v1.4.0

Oct 3, 2025

Fixture patterns