tdd

Name: tdd
Author: eduwxyz

15stars1forks

How to install tdd?

eduwxyz/my-awesome-skills/tdd

$npx -y skills add eduwxyz/my-awesome-skills --skill tdd

Installs into the current project.

Summary

Drives feature work and bug fixes through a tight failing-test-first loop. Trigger when implementing, fixing, or refactoring behavior in a codebase that already has tests. Skip for spikes, visual-only edits, throwaway scripts, and generated files.

Files · 1

View on GitHub

SKILL.md

1	# TDD
2
3	One small failing test, just enough code to pass, then a look. Repeat.
4
5	## Install
6
7	Drop in `~/.claude/skills/tdd/` (per-user) or `<repo>/.claude/skills/tdd/` (per-project).
8
9	## Skip TDD for
10
11	Spikes, copy/style/visual edits, one-off scripts, generated files. If the codebase has no tests at all, see [untested-code.md](untested-code.md) first.
12
13	## Two rules that keep TDD honest
14
15	1. Never start a second test before the first one passes.
16	2. Never edit production code while red, except to make red green.
17
18	If either slips, back up.
19
20	## What a good test looks like
21
22	The name describes a capability, not a method. Read it out loud — it should sound like something a user or caller would care about.
23
24	```
25	✅ logged_out_user_cannot_publish_a_post
26	✅ schedule_overlap_returns_409
27
28	❌ post_service_calls_repo_save
29	❌ schedule_returns_object_with_status
30	```
31
32	The body has three sections in order — set up, do the thing, check what's observed:
33
34	```
35	test "expired tokens are rejected":
36	token = issue_token(ttl_seconds: 60)
37	advance_clock(seconds: 120)
38
39	result = verify(token)
40
41	assert result.ok == false
42	assert result.reason == "expired"
43	```
44
45	If renaming a private function tomorrow would break the test even though no behavior changed, the test was tied to internals. Rewrite or delete it. See [test-anatomy.md](test-anatomy.md).
46
47	## The cardinal mistake: tests in bulk
48
49	Writing five tests up front and then five implementations produces tests that describe an imagined system. They lock you into the wrong shape and stop pulling their weight once any pair shares a code path.
50
51	Each test must exist because of something you learned writing the previous one.
52
53	```
54	Wrong: RED t1 t2 t3 t4 t5 → GREEN c1 c2 c3 c4 c5
55	Right: t1→c1, t2→c2, t3→c3, ...
56	```
57
58	Same mistake in miniature: writing one test and reaching inside to "also handle" something it doesn't cover. Don't.
59
60	## Before the cycle — wire up the test command
61
62	The Stop hook runs your tests at the end of every turn (green allows the turn to end, red blocks it). It reads the command from `.agents/tdd/test-command.txt` at the project root.
63
64	If that file is missing, set it up before starting the cycle:
65
66	1. Read `CLAUDE.md` first — projects often document the canonical test command there.
67	2. Otherwise check `package.json` (`scripts.test`), `pyproject.toml` / `pytest.ini`, `Makefile` (`test:` target), `Cargo.toml`, or analogous config files for the project's stack.
68	3. If multiple plausible commands exist (e.g. `test:unit` vs `test:e2e`), ask the user which one to use for TDD.
69
70	Once decided, create `.agents/tdd/` if missing and write the command on a single line to `.agents/tdd/test-command.txt`. Confirm with the user the first time.
71
72	The user can edit the file at any time to change the command (e.g. switching from `npm test` to `npm run test:unit`).
73
74	## The cycle
75
76	1. Decide. If a `spec/<slug>.md` exists for this task, read it — Behaviors + Acceptance criteria are your queue. Otherwise list the behaviors to verify in rough order with whoever cares.
77
78	2. Smoke run. Pick the first behavior. Test → fails → minimum code → passes. If this first cycle takes more than 15 minutes, the slice is too big.
79
80	3. Each next behavior. Same shape. One test at a time. No speculation. No "while I'm here." If a rule slips, see [smells.md](smells.md#warning-signs-while-you-work).
81
82	4. Cleanup. Tests green? Invoke the `simplify` skill on the recent changes — it reviews code for reuse, quality, and efficiency and applies any improvements found. After simplify, see [cleanup.md](cleanup.md) for additional cleanup. Cleanup only on green; never add behavior during cleanup. (The Stop hook will also remind you to invoke `simplify` once per session if you forget.)
83
84	## When you're done
85
86	Stop adding tests when all are true:
87
88	- Every acceptance criterion has a test.
89	- Every interesting branch has a test (skip getters, plumbing).
90	- The domain edge cases (empty input, expired things, races, boundaries) are covered.
91	- Reading just the test file would teach the feature.
92
93	100% coverage with bad tests is worse than 70% with good ones.
94
95	## Per-cycle checklist
96
97	- [ ] Test names a capability, not a method.
98	- [ ] Test only touches the public API.
99	- [ ] Test would survive a rename of internals.
100	- [ ] Code written is the minimum for this test.
101	- [ ] No second test queued before this one is green.
102
103	## Map
104
105	- [test-anatomy.md](test-anatomy.md) — body shape, assertions per test, naming.
106	- [smells.md](smells.md) — bad-test patterns + warning signs while you work or review.
107	- [boundaries.md]

People who liked this also viewed

6 picks

Type

tdd

15stars1forks

View on GitHub

How to install tdd?

eduwxyz/my-awesome-skills/tdd

$npx -y skills add eduwxyz/my-awesome-skills --skill tdd

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# TDD
2
3	One small failing test, just enough code to pass, then a look. Repeat.
4
5	## Install
6
7	Drop in `~/.claude/skills/tdd/` (per-user) or `<repo>/.claude/skills/tdd/` (per-project).
8
9	## Skip TDD for
10
11	Spikes, copy/style/visual edits, one-off scripts, generated files. If the codebase has no tests at all, see [untested-code.md](untested-code.md) first.
12
13	## Two rules that keep TDD honest
14
15	1. Never start a second test before the first one passes.
16	2. Never edit production code while red, except to make red green.
17
18	If either slips, back up.
19
20	## What a good test looks like
21
22	The name describes a capability, not a method. Read it out loud — it should sound like something a user or caller would care about.
23
24	```
25	✅ logged_out_user_cannot_publish_a_post
26	✅ schedule_overlap_returns_409
27
28	❌ post_service_calls_repo_save
29	❌ schedule_returns_object_with_status
30	```
31
32	The body has three sections in order — set up, do the thing, check what's observed:
33
34	```
35	test "expired tokens are rejected":
36	token = issue_token(ttl_seconds: 60)
37	advance_clock(seconds: 120)
38
39	result = verify(token)
40
41	assert result.ok == false
42	assert result.reason == "expired"
43	```
44
45	If renaming a private function tomorrow would break the test even though no behavior changed, the test was tied to internals. Rewrite or delete it. See [test-anatomy.md](test-anatomy.md).
46
47	## The cardinal mistake: tests in bulk
48
49	Writing five tests up front and then five implementations produces tests that describe an imagined system. They lock you into the wrong shape and stop pulling their weight once any pair shares a code path.
50
51	Each test must exist because of something you learned writing the previous one.
52
53	```
54	Wrong: RED t1 t2 t3 t4 t5 → GREEN c1 c2 c3 c4 c5
55	Right: t1→c1, t2→c2, t3→c3, ...
56	```
57
58	Same mistake in miniature: writing one test and reaching inside to "also handle" something it doesn't cover. Don't.
59
60	## Before the cycle — wire up the test command
61
62	The Stop hook runs your tests at the end of every turn (green allows the turn to end, red blocks it). It reads the command from `.agents/tdd/test-command.txt` at the project root.
63
64	If that file is missing, set it up before starting the cycle:
65
66	1. Read `CLAUDE.md` first — projects often document the canonical test command there.
67	2. Otherwise check `package.json` (`scripts.test`), `pyproject.toml` / `pytest.ini`, `Makefile` (`test:` target), `Cargo.toml`, or analogous config files for the project's stack.
68	3. If multiple plausible commands exist (e.g. `test:unit` vs `test:e2e`), ask the user which one to use for TDD.
69
70	Once decided, create `.agents/tdd/` if missing and write the command on a single line to `.agents/tdd/test-command.txt`. Confirm with the user the first time.
71
72	The user can edit the file at any time to change the command (e.g. switching from `npm test` to `npm run test:unit`).
73
74	## The cycle
75
76	1. Decide. If a `spec/<slug>.md` exists for this task, read it — Behaviors + Acceptance criteria are your queue. Otherwise list the behaviors to verify in rough order with whoever cares.
77
78	2. Smoke run. Pick the first behavior. Test → fails → minimum code → passes. If this first cycle takes more than 15 minutes, the slice is too big.
79
80	3. Each next behavior. Same shape. One test at a time. No speculation. No "while I'm here." If a rule slips, see [smells.md](smells.md#warning-signs-while-you-work).
81
82	4. Cleanup. Tests green? Invoke the `simplify` skill on the recent changes — it reviews code for reuse, quality, and efficiency and applies any improvements found. After simplify, see [cleanup.md](cleanup.md) for additional cleanup. Cleanup only on green; never add behavior during cleanup. (The Stop hook will also remind you to invoke `simplify` once per session if you forget.)
83
84	## When you're done
85
86	Stop adding tests when all are true:
87
88	- Every acceptance criterion has a test.
89	- Every interesting branch has a test (skip getters, plumbing).
90	- The domain edge cases (empty input, expired things, races, boundaries) are covered.
91	- Reading just the test file would teach the feature.
92
93	100% coverage with bad tests is worse than 70% with good ones.
94
95	## Per-cycle checklist
96
97	- [ ] Test names a capability, not a method.
98	- [ ] Test only touches the public API.
99	- [ ] Test would survive a rename of internals.
100	- [ ] Code written is the minimum for this test.
101	- [ ] No second test queued before this one is green.
102
103	## Map
104
105	- [test-anatomy.md](test-anatomy.md) — body shape, assertions per test, naming.
106	- [smells.md](smells.md) — bad-test patterns + warning signs while you work or review.
107	- [boundaries.md]

People who liked this also viewed

6 picks

Type