Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Keep a Mac awake with caffeinate during long builds, downloads, or supervised automation runs.
.claude/skills/sickn33-anti-sleep/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-12 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 34% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-10 | ✓→✗ | ▼ Worse | -57% | 0% |
| case-02 | ✓→✓ | = Same ✓ | 14% | 0% |
Keep the Mac awake using the built-in caffeinate command. No install needed.
bashcaffeinate -d -i -t 7200 # full power: screen stays on + no idle sleep, for 2 hours
Duration is -t <seconds>: 2h = 7200, 7h = 25200, overnight (9h) = 32400.
| Flags | Effect | |---|---| | -i | prevents idle system sleep only (screen may still dim/lock) | | -d | prevents display sleep (screen stays on) | | -d -i | default choice — screen on + system awake | | -d -i -s | adds -s: prevents sleep even on AC power semantics; -s only works when plugged in | | -u -t 1 | simulates user activity — wakes the display right now |
Default to -d -i -t <seconds> unless the user says otherwise.
bashcaffeinate -d -i -w <PID> # stays awake until that process exits (great for builds) caffeinate -i npm run build # wraps a command; exits when the command finishes
Prefer running it in the user's own terminal pane so it's visible and easy to Ctrl+C. In cmux (read the cmux skill first if interacting with panes):
bashcmux send --surface surface:<N> "caffeinate -d -i -t 25200\n"
Otherwise run it as a background Bash task. Never block your own foreground shell with it.
bashpgrep -fl caffeinate # is it running? shows exact flags ps -o etime= -p <PID> # how long it's been running pmset -g assertions | grep -i deny # confirm sleep assertions are active
Gotcha: caffeinate prints nothing and holds the prompt — it looks "stuck" or like Enter wasn't pressed. It isn't stuck. Verify with pgrep, not by looking at the terminal.
Expiry: with -t it exits silently when time runs out — no notification. If the user asks "is it still on?" after hours, check pgrep first; it may simply have expired.
caffeinate cannot keep the keyboard backlight on — it has its own inactivity timer with no CLI/API on Apple Silicon (researched 2026-07). Fix is manual, one-time: System Settings > Keyboard > "Turn keyboard backlight off after inactivity" > Never.
bashpkill -f "caffeinate -d -i" # or Ctrl+C in the pane running it
After starting: confirm to the user the PID, the flags, and the wall-clock time it will expire.
davidondrej/skills; verify local paths, tools, credentials, and agent features before acting.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 4,964 | 6,404 | +29% | 1 | 1 | 0% | 764 | 1,162 | +52% | 0 | 0 | — |
case-02 | pass→pass | 6,559 | 3,202 | -51% | 1 | 1 | 0% | 1,209 | 1,377 | +14% | 0 | 0 | — |
case-03 | pass→pass | 7,666 | 3,501 | -54% | 1 | 1 | 0% | 1,403 | 1,466 | +4% | 0 | 0 | — |
case-04 | pass→pass | 5,348 | 2,652 | -50% | 1 | 1 | 0% | 936 | 1,367 | +46% | 0 | 0 | — |
case-05 | pass→pass | 5,373 | 2,907 | -46% | 1 | 1 | 0% | 1,143 | 1,284 | +12% | 0 | 0 | — |
case-06 | pass→pass | 9,366 | 2,839 | -70% | 1 | 1 | 0% | 1,692 | 1,197 | -29% | 0 | 0 | — |
case-07 | pass→pass | 7,966 | 4,191 | -47% | 1 | 1 | 0% | 1,425 | 1,637 | +15% | 0 | 0 | — |
case-21 | pass→pass | 9,626 | 6,120 | -36% | 1 | 1 | 0% | 1,692 | 1,920 | +13% | 0 | 0 | — |
case-08 | pass→pass | 6,614 | 3,015 | -54% | 1 | 1 | 0% | 1,173 | 1,320 | +13% | 0 | 0 | — |
case-09 | pass→pass | 6,170 | 2,760 | -55% | 1 | 1 | 0% | 1,072 | 1,347 | +26% | 0 | 0 | — |
case-10 | pass→fail | 13,895 | 2,226 | -84% | 1 | 1 | 0% | 2,421 | 1,034 | -57% | 0 | 0 | — |
case-11 | pass→pass | 8,270 | 1,960 | -76% | 1 | 1 | 0% | 1,416 | 1,124 | -21% | 0 | 0 | — |
case-22 | pass→pass | 9,954 | 5,991 | -40% | 1 | 1 | 0% | 1,789 | 1,827 | +2% | 0 | 0 | — |
case-12 | fail→pass | 5,344 | 3,235 | -39% | 1 | 1 | 0% | 1,156 | 1,406 | +22% | 0 | 0 | — |
case-13 | pass→pass | 11,612 | 2,955 | -75% | 1 | 1 | 0% | 1,827 | 1,412 | -23% | 0 | 0 | — |
case-14 | pass→pass | 8,489 | 1,464 | -83% | 1 | 1 | 0% | 1,568 | 988 | -37% | 0 | 0 | — |
case-15 | pass→pass | 11,441 | 1,970 | -83% | 1 | 1 | 0% | 1,924 | 1,137 | -41% | 0 | 0 | — |
case-16 | pass→pass | 5,388 | 3,183 | -41% | 1 | 1 | 0% | 955 | 1,376 | +44% | 0 | 0 | — |
case-17 | fail→pass | 6,162 | 4,187 | -32% | 1 | 1 | 0% | 1,121 | 1,498 | +34% | 0 | 0 | — |
case-18 | pass→pass | 8,106 | 4,882 | -40% | 1 | 1 | 0% | 1,507 | 1,580 | +5% | 0 | 0 | — |
case-19 | pass→pass | 3,588 | 2,189 | -39% | 1 | 1 | 0% | 541 | 1,151 | +113% | 0 | 0 | — |
case-20 | fail→pass | 15,086 | 5,752 | -62% | 1 | 1 | 0% | 2,552 | 1,812 | -29% | 0 | 0 | — |
case-23 | pass→pass | 12,041 | 10,200 | -15% | 1 | 1 | 0% | 2,296 | 2,939 | +28% | 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 22 counted toward the lift figure. The other 1 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 +9 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 cases got worse with the skill loaded, and they are 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.