CtrlK
BlogDocsLog inGet started
Tessl Logo

tdd

Strict TDD loop — red/green/refactor must be followed for every code change, no exceptions

74

1.42x
Quality

61%

Does it follow best practices?

Impact

94%

1.42x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./dot_config/opencode/skill/tdd/SKILL.md
SKILL.md
Quality
Evals
Security

Evaluation results

96%

49%

Email Address Validator Library

Red/green/refactor cycle documentation

Criteria
Without context
With context

Checkpoint format: Red

0%

100%

Checkpoint format: Green

0%

100%

Checkpoint format: Refactor

0%

100%

Checkpoint format: Committed

0%

100%

Multiple cycles logged

70%

100%

Failure confirmed before implementation

83%

100%

Behavior-focused commit messages

83%

66%

Multiple commits (one per cycle)

20%

100%

Minimal implementation

83%

100%

Test-first evidence

80%

100%

87%

21%

Temperature Conversion Library

One cycle per feature, commit discipline

Criteria
Without context
With context

Separate cycle per conversion

100%

100%

Commit per cycle

0%

66%

Checkpoint Red entries present

70%

100%

Checkpoint Green entries present

62%

100%

Checkpoint Refactor entries present

71%

100%

Checkpoint Committed entries present

28%

100%

No behavior added in refactor

100%

100%

Behavior-focused commit messages

53%

38%

Tests written before implementation

100%

100%

100%

14%

Shopping Cart Discount Calculator — Add Bulk Discount Feature

TDD recovery from untested code

Criteria
Without context
With context

TDD gap documented

100%

100%

Tests written for existing code

100%

100%

Existing tests pass

100%

100%

New feature test fails first

100%

100%

Red entry has failure detail

40%

100%

Checkpoint Green entry for new feature

100%

100%

Checkpoint Refactor entry for new feature

0%

100%

Checkpoint Committed entry for new feature

100%

100%

Minimal new implementation

100%

100%

Behavior-focused commit message

70%

100%

Repository
athal7/dotfiles
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.