Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Experimental Claude Code skills extending capabilities with new techniques for interactive CLI tools, tmux automation, and advanced workflows.
.claude/skills/microck-superpowers-lab/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -30% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -25% | 0% |
Experimental skills that extend Claude Code with innovative techniques and tools under active development. These skills enable new automation patterns and capabilities.
Enables Claude Code to control interactive CLI tools through tmux sessions for automation of traditionally manual workflows.
Supported Tools:
How It Works:
See skills/using-tmux-for-interactive-commands/SKILL.md for detailed documentation.
Skills in this collection are:
Explore the subdirectories for individual skill documentation and implementation details. Each skill is self-contained and can be integrated independently.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | pass→pass | 7,653 | 3,481 | -55% | 1 | 1 | 0% | 1,358 | 889 | -35% | 0 | 0 | — |
case-01 | fail→fail | 8,668 | 7,025 | -19% | 1 | 1 | 0% | 1,511 | 895 | -41% | 0 | 0 | — |
case-02 | fail→fail | 4,481 | 5,945 | +33% | 1 | 1 | 0% | 659 | 611 | -7% | 0 | 0 | — |
case-03 | fail→fail | 19,305 | 30,908 | +60% | 1 | 1 | 0% | 2,725 | 490 | -82% | 0 | 0 | — |
case-04 | fail→pass | 14,833 | 7,879 | -47% | 1 | 1 | 0% | 2,514 | 1,750 | -30% | 0 | 0 | — |
case-05 | fail→pass | 12,858 | 6,877 | -47% | 1 | 1 | 0% | 2,335 | 1,483 | -36% | 0 | 0 | — |
case-06 | pass→pass | 13,224 | 8,637 | -35% | 1 | 1 | 0% | 2,411 | 1,903 | -21% | 0 | 0 | — |
case-07 | fail→pass | 14,134 | 9,212 | -35% | 1 | 1 | 0% | 2,423 | 1,904 | -21% | 0 | 0 | — |
case-08 | fail→pass | 13,895 | 8,975 | -35% | 1 | 1 | 0% | 2,465 | 2,107 | -15% | 0 | 0 | — |
case-09 | fail→pass | 14,177 | 9,135 | -36% | 1 | 1 | 0% | 2,366 | 1,776 | -25% | 0 | 0 | — |
case-10 | pass→pass | 13,032 | 8,493 | -35% | 1 | 1 | 0% | 2,318 | 1,711 | -26% | 0 | 0 | — |
case-11 | pass→pass | 6,917 | 3,864 | -44% | 1 | 1 | 0% | 1,120 | 932 | -17% | 0 | 0 | — |
case-12 | fail→pass | 9,852 | 4,672 | -53% | 1 | 1 | 0% | 1,503 | 1,018 | -32% | 0 | 0 | — |
case-13 | pass→pass | 11,979 | 1,343 | -89% | 1 | 1 | 0% | 1,911 | 449 | -77% | 0 | 0 | — |
case-14 | pass→pass | 11,733 | 4,353 | -63% | 1 | 1 | 0% | 1,811 | 1,014 | -44% | 0 | 0 | — |
case-15 | pass→pass | 9,204 | 3,999 | -57% | 1 | 1 | 0% | 1,825 | 1,007 | -45% | 0 | 0 | — |
case-16 | pass→pass | 12,668 | 5,504 | -57% | 1 | 1 | 0% | 2,344 | 1,306 | -44% | 0 | 0 | — |
case-17 | pass→pass | 11,696 | 5,060 | -57% | 1 | 1 | 0% | 2,163 | 1,182 | -45% | 0 | 0 | — |
case-18 | pass→pass | 11,842 | 4,685 | -60% | 1 | 1 | 0% | 1,839 | 1,007 | -45% | 0 | 0 | — |
case-19 | pass→pass | 7,248 | 2,280 | -69% | 1 | 1 | 0% | 1,184 | 668 | -44% | 0 | 0 | — |
case-21 | pass→pass | 13,104 | 7,989 | -39% | 1 | 1 | 0% | 2,114 | 1,549 | -27% | 0 | 0 | — |
case-22 | fail→fail | 9,444 | 6,096 | -35% | 1 | 1 | 0% | 1,649 | 1,433 | -13% | 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 19 counted toward the lift figure. The other 3 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 +27 percentage points is the difference between those two pass rates over the 19 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.