Read the test. Try the question.
tdd
The prompts and scoring guides used to test tdd, with each report’s findings and limits.
mattpocock/skills
Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.
The task
test
Link to this taskWritten responses only. No tool execution, installed dependencies, media generation or cloud services are tested.
What we asked
Do not use tools, access the network, write files, run commands or install anything. Answer only in text using the supplied facts. The agreed public seam is reserve(stock, amount), returning remaining stock. For stock 4 and amount 2, the required result is 2. Draft the first test and the next step in a test-first cycle.
The scoring guide
Pass threshold: 0.75.
Driftproof-authored from the pinned SKILL.md and README; no separate drafting model call.
Try it yourself
Paste the prompt into your own assistant, with or without the skill. Use the scoring guide to read the answer.
One try scored by hand is not a measurement.
Report 017: several answers per arm, each scored 5 times by a judge model. The report’s findings and caveats are below.
Report 018: several answers per arm, each scored 5 times by a judge model. The report’s findings and caveats are below.
Copying stays in your browser. Copies are never stored, counted or sent anywhere by Driftproof.
What happened on this task
Open a report to read its comparison and limits together. Beating the fake is not beating no skill.
Claude Sonnet 5.5Report 017 · Read the task result
Against the same-length fake
test (too few answers to tell): with the skill 1 plus or minus 0; with the same-length placebo 0.983333 plus or minus 0.028868.
Against nothing
test (too few answers to tell): with the skill 1 plus or minus 0; without it 0.983333 plus or minus 0.028868.
- What we tested. Report 015's twenty skills on Sonnet 5.5, each against no skill and against a same-length text from a fixed corpus, judged by Haiku 5.5.
- Beating the fake is not beating no skill. Where the fake scored below no skill, beating it does not mean beating no skill; each row shows both readings.
- What it does not show. It does not show how these skills do beyond three text tasks each, written by Driftproof for Report 015's twenty skills from the Skills.sh snapshot of 2026-10-17, on Claude Sonnet 5.5 judged by Claude Haiku 5.5.
Driftproof collected several answers per arm. A judge model scored each answer 5 times.
Claude Haiku 5.5Report 018 · Read the task result
Against the same-length fake
test (too few answers to tell): with the skill 0.983333 plus or minus 0.028868; with the same-length placebo 0.966667 plus or minus 0.028868.
Against nothing
test (too few answers to tell): with the skill 0.983333 plus or minus 0.028868; without it 1 plus or minus 0.
- What we tested. Twenty skills on Haiku 5.5, each against no skill and against a same-length text from a fixed corpus, judged by Haiku 5.5, the same model.
- The judge is the target's model. The judge uses the same model as the target. Self-grading may favor its own model's answers.
- Beating the fake is not beating no skill. Where the fake scored below no skill, beating it does not mean beating no skill; each row shows both readings.
- What it does not show. It does not show how these skills do beyond three text tasks each, written by Driftproof for the twenty skills from the Skills.sh snapshot of 2026-10-17, on Claude Haiku 5.5 judged by Claude Haiku 5.5.
Driftproof collected several answers per arm. A judge model scored each answer 5 times.
The task
coupling
Link to this taskWritten responses only. No tool execution, installed dependencies, media generation or cloud services are tested.
What we asked
Do not use tools, access the network, write files, run commands or install anything. Answer only in text using the supplied facts. Review this test idea: spy on a private normalizeCart helper and assert it is called once, then assert total equals items.reduce with the production formula. The public requirement is a two-item order totaling 15.
The scoring guide
Pass threshold: 0.75.
Driftproof-authored from the pinned SKILL.md and README; no separate drafting model call.
Try it yourself
Paste the prompt into your own assistant, with or without the skill. Use the scoring guide to read the answer.
One try scored by hand is not a measurement.
Report 017: several answers per arm, each scored 5 times by a judge model. The report’s findings and caveats are below.
Report 018: several answers per arm, each scored 5 times by a judge model. The report’s findings and caveats are below.
Copying stays in your browser. Copies are never stored, counted or sent anywhere by Driftproof.
What happened on this task
Open a report to read its comparison and limits together. Beating the fake is not beating no skill.
Claude Sonnet 5.5Report 017 · Read the task result
Against the same-length fake
coupling (too few answers to tell): with the skill 1 plus or minus 0; with the same-length placebo 0.983333 plus or minus 0.028868.
Against nothing
coupling (too few answers to tell): with the skill 1 plus or minus 0; without it 0.966667 plus or minus 0.028868.
- What we tested. Report 015's twenty skills on Sonnet 5.5, each against no skill and against a same-length text from a fixed corpus, judged by Haiku 5.5.
- Beating the fake is not beating no skill. Where the fake scored below no skill, beating it does not mean beating no skill; each row shows both readings.
- What it does not show. It does not show how these skills do beyond three text tasks each, written by Driftproof for Report 015's twenty skills from the Skills.sh snapshot of 2026-10-17, on Claude Sonnet 5.5 judged by Claude Haiku 5.5.
Driftproof collected several answers per arm. A judge model scored each answer 5 times.
Claude Haiku 5.5Report 018 · Read the task result
Against the same-length fake
coupling (no clear difference): with the skill 1 plus or minus 0; with the same-length placebo 1 plus or minus 0.
Against nothing
coupling (no clear difference): with the skill 1 plus or minus 0; without it 1 plus or minus 0.
- What we tested. Twenty skills on Haiku 5.5, each against no skill and against a same-length text from a fixed corpus, judged by Haiku 5.5, the same model.
- The judge is the target's model. The judge uses the same model as the target. Self-grading may favor its own model's answers.
- Beating the fake is not beating no skill. Where the fake scored below no skill, beating it does not mean beating no skill; each row shows both readings.
- What it does not show. It does not show how these skills do beyond three text tasks each, written by Driftproof for the twenty skills from the Skills.sh snapshot of 2026-10-17, on Claude Haiku 5.5 judged by Claude Haiku 5.5.
Driftproof collected several answers per arm. A judge model scored each answer 5 times.
The task
agreement
Link to this taskWritten responses only. No tool execution, installed dependencies, media generation or cloud services are tested.
What we asked
Do not use tools, access the network, write files, run commands or install anything. Answer only in text using the supplied facts. A teammate asks you to test all billing internals before a new refund feature. No public test seam has been agreed. Explain the next action and one proposed seam with its limits.
The scoring guide
Pass threshold: 0.75.
Driftproof-authored from the pinned SKILL.md and README; no separate drafting model call.
Try it yourself
Paste the prompt into your own assistant, with or without the skill. Use the scoring guide to read the answer.
One try scored by hand is not a measurement.
Report 017: several answers per arm, each scored 5 times by a judge model. The report’s findings and caveats are below.
Report 018: several answers per arm, each scored 5 times by a judge model. The report’s findings and caveats are below.
Copying stays in your browser. Copies are never stored, counted or sent anywhere by Driftproof.
What happened on this task
Open a report to read its comparison and limits together. Beating the fake is not beating no skill.
Claude Sonnet 5.5Report 017 · Read the task result
Against the same-length fake
agreement (too few answers to tell): with the skill 0.966667 plus or minus 0.028868; with the same-length placebo 0.95 plus or minus 0.05.
Against nothing
agreement (too few answers to tell): with the skill 0.966667 plus or minus 0.028868; without it 0.966667 plus or minus 0.028868.
- What we tested. Report 015's twenty skills on Sonnet 5.5, each against no skill and against a same-length text from a fixed corpus, judged by Haiku 5.5.
- Beating the fake is not beating no skill. Where the fake scored below no skill, beating it does not mean beating no skill; each row shows both readings.
- What it does not show. It does not show how these skills do beyond three text tasks each, written by Driftproof for Report 015's twenty skills from the Skills.sh snapshot of 2026-10-17, on Claude Sonnet 5.5 judged by Claude Haiku 5.5.
Driftproof collected several answers per arm. A judge model scored each answer 5 times.
Claude Haiku 5.5Report 018 · Read the task result
Against the same-length fake
agreement (clearly higher): with the skill 0.94 plus or minus 0.054772; with the same-length placebo 0.758333 plus or minus 0.09704.
Against nothing
agreement (too few answers to tell): with the skill 0.94 plus or minus 0.054772; without it 0.866667 plus or minus 0.093095.
- What we tested. Twenty skills on Haiku 5.5, each against no skill and against a same-length text from a fixed corpus, judged by Haiku 5.5, the same model.
- The judge is the target's model. The judge uses the same model as the target. Self-grading may favor its own model's answers.
- Beating the fake is not beating no skill. Where the fake scored below no skill, beating it does not mean beating no skill; each row shows both readings.
- What it does not show. It does not show how these skills do beyond three text tasks each, written by Driftproof for the twenty skills from the Skills.sh snapshot of 2026-10-17, on Claude Haiku 5.5 judged by Claude Haiku 5.5.
Driftproof collected several answers per arm. A judge model scored each answer 5 times.