# Test First

> >

- **Type:** Skill
- **Install:** `agentstack add skill-ao92265-claude-code-playbook-test-first`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [ao92265](https://agentstack.voostack.com/s/ao92265)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [ao92265](https://github.com/ao92265)
- **Source:** https://github.com/ao92265/claude-code-playbook/tree/main/skills/test-first

## Install

```sh
agentstack add skill-ao92265-claude-code-playbook-test-first
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Test-First Development

Enforce the Red-Green-Refactor cycle. Tests define the contract before implementation begins.

## Steps

1. **Understand the requirement:**
   - Clarify what the feature/fix should do
   - Identify inputs, outputs, edge cases, and error conditions
   - Ask clarifying questions if the spec is ambiguous

2. **Write failing tests (RED):**
   - Create test file(s) if they don't exist
   - Write test cases that define expected behavior
   - Include happy path, edge cases, and error cases
   - Tests should be specific and descriptive — test names explain the contract

3. **Confirm tests fail for the right reason (NON-NEGOTIABLE):**
   - Run the test suite: `npm test`, `pytest`, `go test`, etc.
   - Verify failures are "function not found" or "wrong return value" — not syntax errors
   - If tests fail for the wrong reason, fix the test first
   - **Never skip this step.** Tests that go green on first run are tautological or mocked the behavior under test. A test you never watched fail is theater, not verification.

4. **Implement minimum code to pass (GREEN):**
   - Write the simplest code that makes all tests pass
   - Do not optimize or generalize yet
   - Do not add code that isn't required by a test

5. **Confirm all tests pass:**
   - Run the full test suite
   - All new tests must be green
   - All existing tests must still pass (no regressions)

6. **Refactor (REFACTOR):**
   - Clean up implementation while keeping tests green
   - Extract helpers, rename variables, simplify logic
   - Run tests after each refactoring step

7. **Final verification:**
   - Run full test suite one last time
   - Run type checker / linter if applicable
   - Show the user the test results and implementation summary

## Important

- **Never skip the RED step.** If you can't write a failing test first, you don't understand the requirement well enough.
- **Never write implementation before tests exist.** The test defines the contract.
- **Keep tests fast.** Prefer unit tests over integration tests where possible.
- **One behavior per test.** Each test should verify exactly one thing.
- **Test names are documentation.** Use descriptive names: `should_return_404_when_user_not_found`.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [ao92265](https://github.com/ao92265)
- **Source:** [ao92265/claude-code-playbook](https://github.com/ao92265/claude-code-playbook)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-ao92265-claude-code-playbook-test-first
- Seller: https://agentstack.voostack.com/s/ao92265
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.