Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Schedule recurring tasks with notifications. Triggered by "/loop <interval> <prompt>" to set up periodic monitoring.
.claude/skills/x-cmd-loop/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 84% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 25% | 0% |
Use when:
/loop <interval> <prompt>/loop <interval> <prompt>Interval patterns: | Pattern | Cron Expression | |---------|----------------| | Nm (N ≤ 59) | */N * * * * | | Nh (N ≤ 23) | 0 */N * * * | | Nd | 0 0 */N * * |
Round non-dividing intervals to nearest clean value.
Ask user:
Ask before creating:
bashx cron add \ --name "loop-$(date +%Y%m%d%H%M%S)" \ --desc "<description>" \ "<cron-expression>" \ "<full-command>"
Run the command now; do not wait for first cron fire.
Report to user:
x cron rm <name>| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 8,504 | 6,427 | -24% | 1 | 1 | 0% | 1,355 | 1,230 | -9% | 0 | 0 | — |
case-02 | fail→fail | 10,268 | 6,282 | -39% | 1 | 1 | 0% | 1,617 | 1,322 | -18% | 0 | 0 | — |
case-03 | fail→pass | 12,848 | 6,699 | -48% | 1 | 1 | 0% | 1,310 | 1,337 | +2% | 0 | 0 | — |
case-04 | pass→pass | 5,515 | 4,630 | -16% | 1 | 1 | 0% | 793 | 1,007 | +27% | 0 | 0 | — |
case-05 | pass→pass | 11,217 | 5,061 | -55% | 1 | 1 | 0% | 1,399 | 1,139 | -19% | 0 | 0 | — |
case-06 | pass→pass | 3,488 | 7,765 | +123% | 1 | 1 | 0% | 466 | 677 | +45% | 0 | 0 | — |
case-07 | fail→pass | 6,440 | 7,717 | +20% | 1 | 1 | 0% | 835 | 1,538 | +84% | 0 | 0 | — |
case-08 | fail→pass | 9,819 | 7,036 | -28% | 1 | 1 | 0% | 1,164 | 1,488 | +28% | 0 | 0 | — |
case-09 | fail→fail | 10,010 | 3,805 | -62% | 1 | 1 | 0% | 1,228 | 855 | -30% | 0 | 0 | — |
case-10 | pass→pass | 7,747 | 4,021 | -48% | 1 | 1 | 0% | 1,172 | 937 | -20% | 0 | 0 | — |
case-11 | fail→pass | 10,107 | 3,397 | -66% | 1 | 1 | 0% | 1,284 | 844 | -34% | 0 | 0 | — |
case-12 | fail→fail | 6,481 | 8,090 | +25% | 1 | 1 | 0% | 927 | 1,771 | +91% | 0 | 0 | — |
case-13 | fail→pass | 6,878 | 5,048 | -27% | 1 | 1 | 0% | 884 | 1,101 | +25% | 0 | 0 | — |
case-14 | fail→pass | 7,679 | 14,254 | +86% | 1 | 1 | 0% | 1,120 | 1,412 | +26% | 0 | 0 | — |
case-15 | fail→pass | 3,887 | 5,986 | +54% | 1 | 1 | 0% | 398 | 1,260 | +217% | 0 | 0 | — |
case-16 | fail→pass | 12,280 | 3,516 | -71% | 1 | 1 | 0% | 2,134 | 863 | -60% | 0 | 0 | — |
case-17 | fail→pass | 14,033 | 4,084 | -71% | 1 | 1 | 0% | 2,285 | 755 | -67% | 0 | 0 | — |
case-18 | fail→pass | 10,358 | 3,657 | -65% | 1 | 1 | 0% | 1,637 | 915 | -44% | 0 | 0 | — |
case-19 | fail→pass | 6,185 | 2,541 | -59% | 1 | 1 | 0% | 807 | 690 | -14% | 0 | 0 | — |
case-20 | fail→fail | 4,347 | 7,853 | +81% | 1 | 1 | 0% | 476 | 1,455 | +206% | 0 | 0 | — |
case-21 | fail→fail | 5,574 | 5,799 | +4% | 1 | 1 | 0% | 718 | 1,198 | +67% | 0 | 0 | — |
case-22 | fail→pass | 14,263 | 6,325 | -56% | 1 | 1 | 0% | 2,331 | 1,352 | -42% | 0 | 0 | — |
case-23 | fail→pass | 5,049 | 8,929 | +77% | 1 | 1 | 0% | 631 | 1,642 | +160% | 0 | 0 | — |
case-24 | fail→fail | 7,637 | 21,054 | +176% | 1 | 1 | 0% | 1,132 | 1,251 | +11% | 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. 24 cases were attempted. The headline lift of +54 percentage points is the difference between those two pass rates over the 24 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.