test-driven-development

Name: test-driven-development
Author: obra

byobra· 95 skills

Installs

181k

Stars

261k

Forks

23k

How to install test-driven-development?

obra/superpowers/test-driven-development

$npx -y skills add obra/superpowers --skill test-driven-development

Installs into the current project.

›Prefer a prompt? Paste this to your agent

Use this skill

Run `npx skills use "https://github.com/obra/superpowers" --skill "obra/superpowers/test-driven-development"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.

Use the whole pack

Use the skills in "https://github.com/obra/superpowers" that are relevant to the current task. Run `npx skills add "https://github.com/obra/superpowers"` and select the relevant skills, then follow their instructions.

Files · 1

View on GitHub

SKILL.md

1	# Test-Driven Development (TDD)
2
3	## Overview
4
5	Write the test first. Watch it fail. Write minimal code to pass.
6
7	Core principle: If you didn't watch the test fail, you don't know if it tests the right thing.
8
9	Violating the letter of the rules is violating the spirit of the rules.
10
11	## When to Use
12
13	Always:
14	- New features
15	- Bug fixes
16	- Refactoring
17	- Behavior changes
18
19	Exceptions (ask your human partner):
20	- Throwaway prototypes
21	- Generated code
22	- Configuration files
23
24	Thinking "skip TDD just this once"? Stop. That's rationalization.
25
26	## The Iron Law
27
28	```
29	NO PRODUCTION CODE WITHOUT A FAILING TEST FIRST
30	```
31
32	Write code before the test? Delete it. Start over.
33
34	No exceptions:
35	- Don't keep it as "reference"
36	- Don't "adapt" it while writing tests
37	- Don't look at it
38	- Delete means delete
39
40	Implement fresh from tests. Period.
41
42	## Red-Green-Refactor
43
44	```dot
45	digraph tdd_cycle {
46	rankdir=LR;
47	red [label="RED\nWrite failing test", shape=box, style=filled, fillcolor="#ffcccc"];
48	verify_red [label="Verify fails\ncorrectly", shape=diamond];
49	green [label="GREEN\nMinimal code", shape=box, style=filled, fillcolor="#ccffcc"];
50	verify_green [label="Verify passes\nAll green", shape=diamond];
51	refactor [label="REFACTOR\nClean up", shape=box, style=filled, fillcolor="#ccccff"];
52	next [label="Next", shape=ellipse];
53
54	red -> verify_red;
55	verify_red -> green [label="yes"];
56	verify_red -> red [label="wrong\nfailure"];
57	green -> verify_green;
58	verify_green -> refactor [label="yes"];
59	verify_green -> green [label="no"];
60	refactor -> verify_green [label="stay\ngreen"];
61	verify_green -> next;
62	next -> red;
63	}
64	```
65
66	### RED - Write Failing Test
67
68	Write one minimal test showing what should happen.
69
70	<Good>
71	```typescript
72	test('retries failed operations 3 times', async () => {
73	let attempts = 0;
74	const operation = () => {
75	attempts++;
76	if (attempts < 3) throw new Error('fail');
77	return 'success';
78	};
79
80	const result = await retryOperation(operation);
81
82	expect(result).toBe('success');
83	expect(attempts).toBe(3);
84	});
85	```
86	Clear name, tests real behavior, one thing
87	</Good>
88
89	<Bad>
90	```typescript
91	test('retry works', async () => {
92	const mock = jest.fn()
93	.mockRejectedValueOnce(new Error())
94	.mockRejectedValueOnce(new Error())
95	.mockResolvedValueOnce('success');
96	await retryOperation(mock);
97	expect(mock).toHaveBeenCalledTimes(3);
98	});
99	```
100	Vague name, tests mock not code
101	</Bad>
102
103	Requirements:
104	- One behavior
105	- Clear name
106	- Real code (no mocks unless unavoidable)
107
108	### Verify RED - Watch It Fail
109
110	MANDATORY. Never skip.
111
112	```bash
113	npm test path/to/test.test.ts
114	```
115
116	Confirm:
117	- Test fails (not errors)
118	- Failure message is expected
119	- Fails because feature missing (not typos)
120
121	Test passes? You're testing existing behavior. Fix test.
122
123	Test errors? Fix error, re-run until it fails correctly.
124
125	### GREEN - Minimal Code
126
127	Write simplest code to pass the test.
128
129	<Good>
130	```typescript
131	async function retryOperation<T>(fn: () => Promise<T>): Promise<T> {
132	for (let i = 0; i < 3; i++) {
133	try {
134	return await fn();
135	} catch (e) {
136	if (i === 2) throw e;
137	}
138	}
139	throw new Error('unreachable');
140	}
141	```
142	Just enough to pass
143	</Good>
144
145	<Bad>
146	```typescript
147	async function retryOperation<T>(
148	fn: () => Promise<T>,
149	options?: {
150	maxRetries?: number;
151	backoff?: 'linear' \| 'exponential';
152	onRetry?: (attempt: number) => void;
153	}
154	): Promise<T> {
155	// YAGNI
156	}
157	```
158	Over-engineered
159	</Bad>
160
161	Don't add features, refactor other code, or "improve" beyond the test.
162
163	### Verify GREEN - Watch It Pass
164
165	MANDATORY.
166
167	```bash
168	npm test path/to/test.test.ts
169	```
170
171	Confirm:
172	- Test passes
173	- Other tests still pass
174	- Output pristine (no errors, warnings)
175
176	Test fails? Fix code, not test.
177
178	Other tests fail? Fix now.
179
180	### REFACTOR - Clean Up
181
182	After green only:
183	- Remove duplication
184	- Improve names
185	- Extract helpers
186
187	Keep tests green. Don't add behavior.
188
189	### Repeat
190
191	Next failing test for next feature.
192
193	## Good Tests
194
195	\| Quality \| Good \| Bad \|
196	\|---------\|------\|-----\|
197	\| Minimal \| One thing. "and" in name? Split it. \| `test('validates email and domain and whitespace')` \|
198	\| Clear \| Name describes behavior \| `test('test1')` \|
199	\| Shows intent \| Demonstrates desired API \| Obscures what code should do \|
200
201	When writing or changing any test, read [writing-good-tests.md](writing-good-tests.md) for the rules that keep tests honest:
202	- Name the production change that would make the test fail — before writing it
203	- Assert on real behavior, never on mock behavior
204	- Keep test-only code in test utilities, out of production classes
205	- Understand a dependency's side effects before mocking it
206
207	## Common Rationalizations
208
209	\| Excuse \| Reality \|
210	\|--------\|---------\|
211	\| "Too simple to test" \| Simple code breaks. Test takes 30 seconds. \|
212	\| "I'll test after" \| Tests written after pass immediately — which proves nothing. They may test the wrong thing, test the implementation instead of the behavior, or miss the edge case you forgot. You never watched it fail, so you never proved it can catch the bug. Test-first forces that failure. \|
213	\| "Tests after achieve same goals (spirit not ritual)" \| Tests-after answer "what does this do?"; tests-first answer "what should this do?" Tests written after are biased by the code you already wrote — you verify the cases you remembered, not the ones you'd have discovered. Coverage without proof the tests work. \|
214	\| "Already manually tested" \| Manual testing is ad-hoc: no record of what you covered, no way to re-run it when the code changes, easy to forget cases under pressure. "Worked when I tried it" ≠ comprehensive. Automated tests run the same way every time. \|
215	\| "Deleting X hours is wasteful" \| Sunk cost fallacy — that time is already spent either way. The real choice: rewrite with TDD (high confidence) vs. keep it and bolt tests on after (low confidence, likely bugs). Keeping code you can't trust is the waste. \|
216	\| "Keep as reference, write tests first" \| You'll adapt it. That's testing after. Delete means delete. \|
217	\| "Need to explore first" \| Fine. Throw away exploration, start with TDD. \|
218	\| "Test hard = design unclear" \| Listen to test. Hard to test = hard to use. \|
219	\| "TDD will slow me down" \| TDD IS the pragmatic path: catches bugs before commit, prevents regressions, lets you refactor without fear. "Pragmatic" shortcuts mean debugging in production — slower, not faster. \|
220	\| "Manual test faster" \| Manual doesn't prove edge cases. You'll re-test every change. \|
221	\| "Existing code has no tests" \| You're improving it. Add tests for existing code. \|
222
223	## Red Flags - STOP and Start Over
224
225	- Code before test
226	- Test after implementation
227	- Test passes immediately
228	- Can't explain why test failed
229	- Tests added "later"
230	- Rationalizing "just this once"
231	- "I already manually tested it"
232	- "Tests after achieve the same purpose"
233	- "It's about spirit not ritual"
234	- "Keep as reference" or "adapt existing code"
235	- "Already spent X hours, deleting is wasteful"
236	- "TDD is dogmatic, I'm being pragmatic"
237	- "This is different because..."
238
239	All of these mean: Delete code. Start over with TDD.
240
241	## Example: Bug Fix
242
243	Bug: Empty email accepted
244
245	RED
246	```typescript
247	test('rejects empty email', async () => {
248	const result = await submitForm({ email: '' });
249	expect(result.error).toBe('Email required');
250	});
251	```
252
253	Verify RED
254	```bash
255	$ npm test
256	FAIL: expected 'Email required', got undefined
257	```
258
259	GREEN
260	```typescript
261	function submitForm(data: FormData) {
262	if (!data.email?.trim()) {
263	return { error: 'Email required' };
264	}
265	// ...
266	}
267	```
268
269	Verify GREEN
270	```bash
271	$ npm test
272	PASS
273	```
274
275	REFACTOR
276	Extract validation for multiple fields if needed.
277
278	## Verification Checklist
279
280	Before marking work complete:
281
282	- [ ] Every new function/method has a test
283	- [ ] Watched each test fail before implementing
284	- [ ] Each test failed for expected reason (feature missing, not typo)
285	- [ ] Wrote minimal code to pass each test
286	- [ ] All tests pass
287	- [ ] Output pristine (no errors, warnings)
288	- [ ] Tests use real code (mocks only if unavoidable)
289	- [ ] Edge cases and errors covered
290
291	Can't check all boxes? You skipped TDD. Start over.
292
293	## When Stuck
294
295	\| Problem \| Solution \|
296	\|---------\|----------\|
297	\| Don't know how to test \| Write wished-for API. Write assertion first. Ask your human partner. \|
298	\| Test too complicated \| Design too complicated. Simplify interface. \|
299	\| Must mock everything \| Code too coupled. Use dependency injection. \|
300	\| Test setup huge \| Extract helpers. Still complex? Simplify design. \|
301
302	## Debugging Integration
303
304	Bug found? Write failing test reproducing it. Follow TDD cycle. Test proves fix and prevents regression.
305
306	Never fix bugs without a test.
307
308	## Final Rule
309
310	```
311	Production code → test exists and failed first
312	Otherwise → not TDD
313	```
314
315	No exceptions without your human partner's permission.

6 picks

Type

Run `npx skills use "https://github.com/obra/superpowers" --skill "obra/superpowers/test-driven-development"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.

Use the skills in "https://github.com/obra/superpowers" that are relevant to the current task. Run `npx skills add "https://github.com/obra/superpowers"` and select the relevant skills, then follow their instructions.

test-driven-development

How to install test-driven-development?

Related

test-driven-development

How to install test-driven-development?

Related