Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Repo-specific dedupe guidance for warp. Only the categories declared overridable by the core dedupe-issue skill may be specialized here.
.claude/skills/warpdotdev-dedupe-issue-local/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -12% | 0% |
warpThis skill specializes the core dedupe-issue skill (named in the specializes frontmatter field) and is not functional on its own. Before applying its guidance, confirm the parent skill is installed and resolvable at .agents/skills/dedupe-issue/SKILL.md. If it is missing, install it first by copying the skill directory from the source declared in the specializes_source frontmatter field (warpdotdev/oz-for-oss:.agents/skills/dedupe-issue). Then continue with the guidance below.
This file is a companion to the core dedupe-issue skill. It does not redefine the duplicate-detection algorithm, the similarity thresholds, or the output contract. It only specializes the override categories the core skill marks as overridable.
Bug:, Feature:, Request:, [Bug], [Feature], Warp:, and platform tags like [macOS], [Linux], or [Windows] before comparing titles.No known-duplicate clusters have been captured for this repository yet. The weekly update-dedupe loop will propose additions here over time when maintainers repeatedly close issues as duplicates of the same canonical thread.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 18,776 | 8,262 | -56% | 1 | 1 | 0% | 3,034 | 887 | -71% | 0 | 0 | — |
case-02 | fail→pass | 10,278 | 5,790 | -44% | 1 | 1 | 0% | 1,526 | 1,446 | -5% | 0 | 0 | — |
case-03 | fail→pass | 7,689 | 4,751 | -38% | 1 | 1 | 0% | 1,117 | 1,222 | +9% | 0 | 0 | — |
case-04 | pass→pass | 16,989 | 8,768 | -48% | 1 | 1 | 0% | 2,592 | 1,875 | -28% | 0 | 0 | — |
case-05 | fail→pass | 13,931 | 8,734 | -37% | 1 | 1 | 0% | 2,083 | 1,833 | -12% | 0 | 0 | — |
case-06 | pass→pass | 10,478 | 7,075 | -32% | 1 | 1 | 0% | 1,665 | 1,636 | -2% | 0 | 0 | — |
case-07 | pass→pass | 6,952 | 4,987 | -28% | 1 | 1 | 0% | 1,140 | 1,334 | +17% | 0 | 0 | — |
case-08 | pass→pass | 11,625 | 5,731 | -51% | 1 | 1 | 0% | 1,783 | 1,334 | -25% | 0 | 0 | — |
case-09 | pass→pass | 11,328 | 5,224 | -54% | 1 | 1 | 0% | 1,766 | 1,304 | -26% | 0 | 0 | — |
case-10 | pass→pass | 22,741 | 7,844 | -66% | 1 | 1 | 0% | 1,868 | 1,620 | -13% | 0 | 0 | — |
case-11 | pass→pass | 10,393 | 5,964 | -43% | 1 | 1 | 0% | 1,685 | 1,327 | -21% | 0 | 0 | — |
case-12 | pass→pass | 16,313 | 5,177 | -68% | 1 | 1 | 0% | 1,133 | 1,318 | +16% | 0 | 0 | — |
case-13 | pass→pass | 3,784 | 4,673 | +23% | 1 | 1 | 0% | 590 | 1,143 | +94% | 0 | 0 | — |
case-14 | pass→pass | 4,684 | 4,914 | +5% | 1 | 1 | 0% | 644 | 1,194 | +85% | 0 | 0 | — |
case-15 | pass→pass | 4,023 | 3,494 | -13% | 1 | 1 | 0% | 602 | 1,003 | +67% | 0 | 0 | — |
case-16 | pass→pass | 4,788 | 5,059 | +6% | 1 | 1 | 0% | 718 | 1,209 | +68% | 0 | 0 | — |
case-17 | fail→pass | 10,573 | 3,258 | -69% | 1 | 1 | 0% | 1,581 | 905 | -43% | 0 | 0 | — |
case-18 | fail→pass | 6,418 | 3,543 | -45% | 1 | 1 | 0% | 1,080 | 945 | -13% | 0 | 0 | — |
case-19 | fail→pass | 9,052 | 6,122 | -32% | 1 | 1 | 0% | 1,430 | 1,279 | -11% | 0 | 0 | — |
case-20 | pass→pass | 16,603 | 10,251 | -38% | 1 | 1 | 0% | 2,761 | 2,101 | -24% | 0 | 0 | — |
case-21 | fail→fail | 16,297 | 24,592 | +51% | 1 | 1 | 0% | 2,938 | 5,148 | +75% | 0 | 0 | — |
case-22 | pass→pass | 16,295 | 9,019 | -45% | 1 | 1 | 0% | 1,883 | 1,824 | -3% | 0 | 0 | — |
case-23 | fail→fail | 32,385 | 15,392 | -52% | 1 | 1 | 0% | 1,291 | 3,128 | +142% | 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. 23 cases were attempted, and 21 counted toward the lift figure. The other 2 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 +26 percentage points is the difference between those two pass rates over the 21 comparable cases.
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.