Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Convert tasks from tasks.md into GitHub issues. Use after task breakdown to track work items in GitHub project management.
.claude/skills/foryourhealth111-pixel-speckit-taskstoissues/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -48% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-20 | ✓→✗ | ▼ Worse | 6% | 0% |
text$ARGUMENTS
You MUST consider the user input before proceeding (if not empty).
.specify/scripts/powershell/check-prerequisites.ps1 -Json -RequireTasks -IncludeTasks from repo root and parse FEATURE_DIR and AVAILABLE_DOCS list. All paths must be absolute. For single quotes in args like "I'm Groot", use escape syntax: e.g 'I'\''m Groot' (or double-quote if possible: "I'm Groot").bashgit config --get remote.origin.url
> !CAUTION] > ONLY PROCEED TO NEXT STEPS IF THE REMOTE IS A GITHUB URL
> !CAUTION] > UNDER NO CIRCUMSTANCES EVER CREATE ISSUES IN REPOSITORIES THAT DO NOT MATCH THE REMOTE URL
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | fail→fail | 8,014 | 3,556 | -56% | 1 | 1 | 0% | 1,447 | 629 | -57% | 0 | 0 | — |
case-01 | fail→fail | 7,239 | 5,675 | -22% | 1 | 1 | 0% | 1,135 | 467 | -59% | 0 | 0 | — |
case-02 | fail→fail | 8,613 | 5,715 | -34% | 1 | 1 | 0% | 1,391 | 466 | -66% | 0 | 0 | — |
case-03 | fail→fail | 7,718 | 6,751 | -13% | 1 | 1 | 0% | 1,243 | 560 | -55% | 0 | 0 | — |
case-05 | fail→fail | 13,130 | 6,466 | -51% | 1 | 1 | 0% | 2,342 | 630 | -73% | 0 | 0 | — |
case-06 | pass→pass | 11,451 | 3,138 | -73% | 1 | 1 | 0% | 2,066 | 798 | -61% | 0 | 0 | — |
case-07 | fail→fail | 12,833 | 7,083 | -45% | 1 | 1 | 0% | 2,076 | 599 | -71% | 0 | 0 | — |
case-08 | fail→fail | 8,378 | 6,039 | -28% | 1 | 1 | 0% | 1,594 | 764 | -52% | 0 | 0 | — |
case-13 | pass→pass | 6,030 | 1,933 | -68% | 1 | 1 | 0% | 897 | 576 | -36% | 0 | 0 | — |
case-09 | fail→pass | 22,774 | 2,639 | -88% | 1 | 1 | 0% | 1,002 | 737 | -26% | 0 | 0 | — |
case-10 | pass→pass | 6,759 | 1,897 | -72% | 1 | 1 | 0% | 1,039 | 503 | -52% | 0 | 0 | — |
case-11 | fail→fail | 11,628 | 3,101 | -73% | 1 | 1 | 0% | 2,305 | 645 | -72% | 0 | 0 | — |
case-12 | pass→pass | 4,330 | 1,073 | -75% | 1 | 1 | 0% | 690 | 383 | -44% | 0 | 0 | — |
case-14 | pass→pass | 8,587 | 2,220 | -74% | 1 | 1 | 0% | 1,735 | 597 | -66% | 0 | 0 | — |
case-15 | fail→pass | 7,192 | 2,588 | -64% | 1 | 1 | 0% | 1,403 | 784 | -44% | 0 | 0 | — |
case-16 | pass→pass | 8,540 | 3,009 | -65% | 1 | 1 | 0% | 1,538 | 699 | -55% | 0 | 0 | — |
case-17 | fail→pass | 11,101 | 3,856 | -65% | 1 | 1 | 0% | 1,720 | 888 | -48% | 0 | 0 | — |
case-18 | fail→fail | 8,483 | 2,100 | -75% | 1 | 1 | 0% | 1,482 | 578 | -61% | 0 | 0 | — |
case-19 | fail→pass | 6,352 | 3,821 | -40% | 1 | 1 | 0% | 1,121 | 1,034 | -8% | 0 | 0 | — |
case-20 | pass→fail | 4,096 | 7,327 | +79% | 1 | 1 | 0% | 750 | 792 | +6% | 0 | 0 | — |
case-21 | pass→pass | 9,981 | 10,621 | +6% | 1 | 1 | 0% | 1,904 | 2,382 | +25% | 0 | 0 | — |
case-22 | pass→pass | 6,962 | 5,024 | -28% | 1 | 1 | 0% | 1,389 | 1,033 | -26% | 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 15 counted toward the lift figure. The other 7 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 +14 percentage points is the difference between those two pass rates over the 15 comparable cases. 1 case got worse with the skill loaded, and it is 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.