Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →macOS native integrations: screen capture, calendar, reminders, contacts, email (Mail.app), Spotlight search. Use when the user asks about their screen, schedule, to-do list, contacts, or wants to send email on macOS.
.claude/skills/sonichi-macos-tools/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-14 | ✗→✓ | ▲ Improved | 303% | 0% |
| case-04 | ✓→✗ | ▼ Worse | 206% | 0% |
| case-09 | ✓→✓ | ＝ Same ✓ | 42% | 0% |
| case-18 | ✓→✓ | ＝ Same ✓ | -38% | 0% |
| case-22 | ✓→✓ | ＝ Same ✓ | -17% | 0% |
Native macOS integrations via AppleScript. No API keys needed — works on any Mac.
bashbash "$SKILL_DIR/scripts/screen-capture.sh"
Returns path to PNG screenshot. Use the Read tool on the path to view it.
Prefer the google-calendar skill if installed. Fallback to macOS Calendar:
bashpython3 "$SKILL_DIR/scripts/calendar-reader.py" 7 # next 7 days, JSON python3 "$SKILL_DIR/scripts/calendar-reader.py" 1 text # today, plain text
bashpython3 "$SKILL_DIR/scripts/reminders.py" list # all incomplete python3 "$SKILL_DIR/scripts/reminders.py" add "Call Bob" # add reminder python3 "$SKILL_DIR/scripts/reminders.py" add "Fix bug" "2026-03-17" # with due date python3 "$SKILL_DIR/scripts/reminders.py" complete "Call Bob" # mark done
bashpython3 "$SKILL_DIR/scripts/contacts.py" search "Bob" # find by name
Returns name, emails, phones. Use before sending email to resolve names to addresses.
bashpython3 "$SKILL_DIR/scripts/email-sender.py" "to@example.com" "Subject" "Body" python3 "$SKILL_DIR/scripts/email-sender.py" "to@example.com" "Subject" "Body" --draft
Sends via Mail.app. Use --draft to create without sending. Always confirm with user before sending.
bashmdfind "quarterly report" # search by content or filename mdfind -name "resume.pdf" # search by filename only
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 5,562 | 6,451 | +16% | 1 | 1 | 0% | 637 | 1,061 | +67% | 0 | 0 | — |
case-02 | fail→fail | 6,270 | 5,412 | -14% | 1 | 1 | 0% | 1,086 | 901 | -17% | 0 | 0 | — |
case-03 | fail→fail | 6,124 | 7,328 | +20% | 1 | 1 | 0% | 952 | 1,156 | +21% | 0 | 0 | — |
case-04 | pass→fail | 5,090 | 16,943 | +233% | 1 | 1 | 0% | 847 | 2,590 | +206% | 0 | 0 | — |
case-05 | fail→fail | 2,976 | 8,184 | +175% | 1 | 1 | 0% | 481 | 967 | +101% | 0 | 0 | — |
case-06 | fail→fail | 2,667 | 10,323 | +287% | 1 | 1 | 0% | 469 | 1,034 | +120% | 0 | 0 | — |
case-07 | fail→fail | 2,534 | 4,772 | +88% | 1 | 1 | 0% | 306 | 840 | +175% | 0 | 0 | — |
case-08 | fail→fail | 8,597 | 5,895 | -31% | 1 | 1 | 0% | 1,151 | 876 | -24% | 0 | 0 | — |
case-09 | pass→pass | 4,916 | 1,439 | -71% | 1 | 1 | 0% | 575 | 819 | +42% | 0 | 0 | — |
case-10 | fail→fail | 3,050 | 6,566 | +115% | 1 | 1 | 0% | 386 | 830 | +115% | 0 | 0 | — |
case-11 | fail→fail | 3,412 | 6,455 | +89% | 1 | 1 | 0% | 484 | 1,086 | +124% | 0 | 0 | — |
case-12 | fail→fail | 2,788 | 5,189 | +86% | 1 | 1 | 0% | 407 | 839 | +106% | 0 | 0 | — |
case-13 | fail→fail | 2,377 | 6,500 | +173% | 1 | 1 | 0% | 322 | 1,127 | +250% | 0 | 0 | — |
case-14 | fail→pass | 2,458 | 4,205 | +71% | 1 | 1 | 0% | 326 | 1,314 | +303% | 0 | 0 | — |
case-15 | fail→fail | 5,601 | 7,486 | +34% | 1 | 1 | 0% | 801 | 1,004 | +25% | 0 | 0 | — |
case-16 | fail→fail | 8,473 | 2,115 | -75% | 1 | 1 | 0% | 1,276 | 905 | -29% | 0 | 0 | — |
case-17 | fail→fail | 6,314 | 4,476 | -29% | 1 | 1 | 0% | 1,049 | 766 | -27% | 0 | 0 | — |
case-18 | pass→pass | 7,799 | 3,282 | -58% | 1 | 1 | 0% | 1,283 | 796 | -38% | 0 | 0 | — |
case-19 | fail→fail | 4,447 | 9,239 | +108% | 1 | 1 | 0% | 677 | 883 | +30% | 0 | 0 | — |
case-20 | fail→fail | 5,427 | 7,395 | +36% | 1 | 1 | 0% | 935 | 993 | +6% | 0 | 0 | — |
case-21 | fail→fail | 1,309 | 6,654 | +408% | 1 | 1 | 0% | 173 | 1,042 | +502% | 0 | 0 | — |
case-22 | pass→pass | 5,325 | 1,703 | -68% | 1 | 1 | 0% | 905 | 747 | -17% | 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 5 counted toward the lift figure. The other 17 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 0 percentage points is the difference between those two pass rates over the 5 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.