test-driven-development

SkillDev tools

Makes your agent write failing tests before writing any code, then implement just enough code to pass them.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the test-driven-development skill

About this capability

Strict RED-GREEN-REFACTOR cycle enforcement. Tests are never skipped or deferred. Run mode only, never watch mode. Exit code evidence mandatory.

What this skill tells your AI

The instructions your AI receives, as published by a5c-ai/babysitter in library/methodologies/cc10x/skills/test-driven-development/SKILL.md and read by ahel’s review.

Non-Negotiable Rules

  • Always use run mode (CI=true npm test or --run flag)
  • NEVER use watch mode (prevents hanging processes)
  • Use timeout guards (timeout 60s) as backup
  • Record exit codes as evidence at each phase
  • If tests fail 3 consecutive times in GREEN, report failure status
  • Never skip RED phase (failing test must exist before implementation)
  • Never defer tests to "later" -- TDD means tests first

Scope Discipline

  • If implementation requires >3 file changes, flag for scope review
  • Architectural choices require user approval unless pre-approved in plan
  • New dependencies require user approval

When to Use

  • Every feature implementation in BUILD workflow
  • Every bug fix in DEBUG workflow (regression test)

Agents Used

  • component-builder (primary consumer)
  • bug-investigator (regression tests)

Signals

GitHub stars
2k
Forks
106
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
test-driven-development-a5c-ai
Source
github.com/a5c-ai/babysitter