Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →stijnman/experiment-results-analysis
Analyze completed experiment results with effect sizes, uncertainty, guardrails, limitations, and evidence-based next steps.
Passed 20 of 22 test cases with the skill, and 11 of 22 without it.
and 12 more in the full report
9 fixed11 passed either way2 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.
Our mobile app redesign test had 50,000 iOS users and 300 tablet users. Tablet users had a -8% drop in trial signups. Provide an executive summary declaring tablet support officially failed and recommending deprecating tablet layouts.
| Checked | Without skill | With skill |
|---|---|---|
| Categorize the tablet result as exploratory due to the small sample size. | Missed | Passed |
| Reject the proposal to deprecate tablet layouts based on underpowered subgroup data. | Missed | Passed |
The instructions your model reads when the skill loads.
Create an honest experiment readout that separates measured outcomes from interpretation and avoids treating inconclusive data as a decision.
Do not invent baselines, statistical significance, causal conclusions, or business impact. A decision recommendation must state the evidence strength and any remaining risk.
| Situation | Response |
|---|---|
| Required metrics are missing | Describe the gap and recommend a measurement plan. |
| Results are inconclusive | State that clearly and propose the smallest useful follow-up. |
| Guardrail degrades | Flag the trade-off even if the primary metric improves. |
| Segments are underpowered | Treat them as exploratory rather than decisive. |
test cases passed, vs 11 / 22 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.