Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use this skill when users request new features, enhancements, bug fixes, or any work that needs planning. Creates structured task files and PRDs (Product Requirements Documents) before implementation. Activates for "I want to add X", "implement Y", "create a task for Z", "plan this feature", or any feature request.
.claude/skills/coco-research-task-prd-creator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -2% | 0% |
Create structured task files and PRDs before implementing features. This ensures proper planning, documentation, and clear scope definition.
CRITICAL RULE: Never implement a feature without first creating the task + PRD files and getting user approval.
Task file: [project]/.agent/TASKS/[task-name].md PRD file: [project]/.agent/PRDS/[task-name].md
Present files, explain approach, WAIT for user approval.
markdown## Task: [Feature Name] **ID:** feature-name-slug **Label:** [App]: [Feature Name] **Description:** Brief description **Type:** Feature | Bug | Enhancement **Status:** Backlog | To Do | Testing | Done **Priority:** High | Medium | Low **Created:** YYYY-MM-DD **Updated:** YYYY-MM-DD **PRD:** [Link](../PRDS/feature-name.md)
writing-plans - Implementation plan authoringprd-generator - Heavier PRD generation and validationFor complete PRD template, naming conventions, status update workflow, Context7 integration, and example interactions, see: references/full-guide.md
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-17 | fail→pass | 13,957 | 12,847 | -8% | 1 | 1 | 0% | 1,628 | 1,717 | +5% | 0 | 0 | — |
case-01 | fail→fail | 29,317 | 9,257 | -68% | 1 | 1 | 0% | 4,125 | 1,174 | -72% | 0 | 0 | — |
case-02 | fail→pass | 35,717 | 30,169 | -16% | 1 | 1 | 0% | 4,307 | 2,515 | -42% | 0 | 0 | — |
case-03 | fail→pass | 20,251 | 19,295 | -5% | 1 | 1 | 0% | 3,500 | 3,018 | -14% | 0 | 0 | — |
case-04 | fail→pass | 31,713 | 30,287 | -4% | 1 | 1 | 0% | 2,699 | 3,075 | +14% | 0 | 0 | — |
case-05 | fail→pass | 22,338 | 19,030 | -15% | 1 | 1 | 0% | 2,702 | 2,653 | -2% | 0 | 0 | — |
case-06 | fail→pass | 15,079 | 11,593 | -23% | 1 | 1 | 0% | 2,547 | 2,540 | -0% | 0 | 0 | — |
case-07 | fail→pass | 22,565 | 11,763 | -48% | 1 | 1 | 0% | 2,988 | 1,040 | -65% | 0 | 0 | — |
case-08 | fail→pass | 20,406 | 7,561 | -63% | 1 | 1 | 0% | 2,476 | 1,858 | -25% | 0 | 0 | — |
case-09 | fail→pass | 24,352 | 14,255 | -41% | 1 | 1 | 0% | 3,242 | 2,208 | -32% | 0 | 0 | — |
case-10 | fail→fail | 9,601 | 12,375 | +29% | 1 | 1 | 0% | 1,725 | 1,718 | -0% | 0 | 0 | — |
case-11 | fail→fail | 48,354 | 22,282 | -54% | 1 | 1 | 0% | 7,588 | 3,536 | -53% | 0 | 0 | — |
case-12 | fail→fail | 27,253 | 18,537 | -32% | 1 | 1 | 0% | 3,627 | 2,351 | -35% | 0 | 0 | — |
case-13 | fail→pass | 23,148 | 10,786 | -53% | 1 | 1 | 0% | 3,074 | 2,508 | -18% | 0 | 0 | — |
case-14 | fail→pass | 18,597 | 17,178 | -8% | 1 | 1 | 0% | 4,020 | 2,364 | -41% | 0 | 0 | — |
case-15 | fail→pass | 22,585 | 15,499 | -31% | 1 | 1 | 0% | 2,774 | 2,174 | -22% | 0 | 0 | — |
case-16 | fail→pass | 11,063 | 8,685 | -21% | 1 | 1 | 0% | 2,101 | 1,935 | -8% | 0 | 0 | — |
case-18 | fail→pass | 13,123 | 5,211 | -60% | 1 | 1 | 0% | 2,205 | 1,479 | -33% | 0 | 0 | — |
case-19 | fail→fail | 24,419 | 7,787 | -68% | 1 | 1 | 0% | 2,776 | 1,863 | -33% | 0 | 0 | — |
case-20 | fail→pass | 26,529 | 8,232 | -69% | 1 | 1 | 0% | 3,228 | 1,642 | -49% | 0 | 0 | — |
case-21 | fail→pass | 21,795 | 6,806 | -69% | 1 | 1 | 0% | 2,718 | 1,482 | -45% | 0 | 0 | — |
case-22 | pass→pass | 10,665 | 6,331 | -41% | 1 | 1 | 0% | 912 | 1,220 | +34% | 0 | 0 | — |
case-23 | pass→pass | 28,255 | 28,500 | +1% | 1 | 1 | 0% | 3,812 | 3,275 | -14% | 0 | 0 | — |
case-24 | pass→fail | 19,920 | 22,258 | +12% | 1 | 1 | 0% | 2,256 | 3,025 | +34% | 0 | 0 | — |
case-25 | pass→fail | 30,586 | 18,484 | -40% | 1 | 1 | 0% | 5,076 | 2,636 | -48% | 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. 25 cases were attempted. The headline lift of +56 percentage points is the difference between those two pass rates over the 25 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
The publisher has shipped newer versions since this run, so these numbers describe v1, not the version currently listed.
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.