Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Turn the current conversation context into a PRD and submit it as a GitHub issue. Use when user wants to create a PRD from the current context.
.claude/skills/mkurman-to-prd/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-18 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-04 | ✓→✗ | ▼ Worse | -68% | 0% |
| case-06 | ✓→✗ | ▼ Worse | -69% | 0% |
| case-07 | ✓→✗ | ▼ Worse | -69% | 0% |
| case-09 | ✓→✗ | ▼ Worse | -86% | 0% |
This skill takes the current conversation context and codebase understanding and produces a PRD. Do NOT interview the user — just synthesize what you already know.
A deep module (as opposed to a shallow module) is one which encapsulates a lot of functionality in a simple, testable interface which rarely changes.
Check with the user that these modules match their expectations. Check with the user which modules they want tests written for.
<prd-template>
The problem that the user is facing, from the user's perspective.
The solution to the problem, from the user's perspective.
A LONG, numbered list of user stories. Each user story should be in the format of:
<user-story-example>
</user-story-example>
This list of user stories should be extremely extensive and cover all aspects of the feature.
A list of implementation decisions that were made. This can include:
Do NOT include specific file paths or code snippets. They may end up being outdated very quickly.
A list of testing decisions that were made. Include:
A description of the things that are out of scope for this PRD.
Any further notes about the feature.
</prd-template>
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-12 | fail→fail | 12,164 | 4,191 | -66% | 1 | 1 | 0% | 2,114 | 716 | -66% | 0 | 0 | — |
case-01 | fail→fail | 5,937 | 3,708 | -38% | 1 | 1 | 0% | 392 | 682 | +74% | 0 | 0 | — |
case-02 | fail→fail | 11,979 | 4,566 | -62% | 1 | 1 | 0% | 1,853 | 820 | -56% | 0 | 0 | — |
case-03 | fail→fail | 19,302 | 2,586 | -87% | 1 | 1 | 0% | 3,589 | 623 | -83% | 0 | 0 | — |
case-04 | pass→fail | 15,529 | 6,438 | -59% | 1 | 1 | 0% | 2,614 | 830 | -68% | 0 | 0 | — |
case-05 | fail→fail | 5,160 | 5,115 | -1% | 1 | 1 | 0% | 759 | 967 | +27% | 0 | 0 | — |
case-06 | pass→fail | 14,105 | 4,073 | -71% | 1 | 1 | 0% | 2,291 | 721 | -69% | 0 | 0 | — |
case-07 | pass→fail | 16,963 | 5,138 | -70% | 1 | 1 | 0% | 2,734 | 834 | -69% | 0 | 0 | — |
case-08 | fail→fail | 13,749 | 4,093 | -70% | 1 | 1 | 0% | 2,242 | 606 | -73% | 0 | 0 | — |
case-09 | pass→fail | 22,512 | 4,833 | -79% | 1 | 1 | 0% | 4,328 | 618 | -86% | 0 | 0 | — |
case-10 | pass→fail | 14,604 | 4,415 | -70% | 1 | 1 | 0% | 2,505 | 688 | -73% | 0 | 0 | — |
case-11 | fail→fail | 8,313 | 4,183 | -50% | 1 | 1 | 0% | 1,397 | 795 | -43% | 0 | 0 | — |
case-13 | pass→fail | 18,036 | 3,537 | -80% | 1 | 1 | 0% | 3,117 | 717 | -77% | 0 | 0 | — |
case-14 | pass→fail | 21,122 | 4,537 | -79% | 1 | 1 | 0% | 3,454 | 787 | -77% | 0 | 0 | — |
case-15 | fail→fail | 20,203 | 4,742 | -77% | 1 | 1 | 0% | 3,308 | 724 | -78% | 0 | 0 | — |
case-16 | fail→fail | 16,508 | 4,609 | -72% | 1 | 1 | 0% | 2,707 | 651 | -76% | 0 | 0 | — |
case-17 | fail→fail | 17,069 | 38,256 | +124% | 1 | 1 | 0% | 2,807 | 6,838 | +144% | 0 | 0 | — |
case-18 | fail→pass | 20,412 | 23,688 | +16% | 1 | 1 | 0% | 3,744 | 4,058 | +8% | 0 | 0 | — |
case-19 | fail→fail | 16,040 | 4,741 | -70% | 1 | 1 | 0% | 2,836 | 828 | -71% | 0 | 0 | — |
case-20 | pass→fail | 4,203 | 5,892 | +40% | 1 | 1 | 0% | 683 | 994 | +46% | 0 | 0 | — |
case-21 | pass→fail | 21,788 | 4,832 | -78% | 1 | 1 | 0% | 4,717 | 762 | -84% | 0 | 0 | — |
case-22 | pass→fail | 16,863 | 3,741 | -78% | 1 | 1 | 0% | 3,704 | 859 | -77% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 1 counted toward the lift figure. The other 21 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of -41 percentage points is the difference between those two pass rates over the 1 comparable cases. 11 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.