Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when the user wants a morning brief, start-of-day summary, routine-ready daily digest, or a concise operational overview combining calendar, tasks, PRs, and issues.
.claude/skills/mkurman-daily-brief/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 82% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 55% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-21 | ✗→✓ | ▲ Improved | -45% | 0% |
I want a concise, source-linked day-start brief that merges my local daemon state with optional inbox and calendar signal, so I can start work without manually cross-checking multiple surfaces.
gmail for urgent unread / follow-up candidatescalendar for today’s agenda and meeting prepmode: standard, quiet, or executiveinclude_inbox: booleaninclude_calendar: booleandelivery_channel: in-app, slack, discord, telegram, or whatsapptime_window: e.g. today / next 8h / next 24hgmail is ready, fetch inbox shortlist.calendar is ready, fetch upcoming events.available vs omitted sections.Use a task-backed routine scheduled for weekday mornings. Example target_payload.description:
Prepare the Daily Brief pack with executive mode, optional Gmail/Calendar enrichment, and mobile-safe delivery summary.
Run the Daily Brief pack in executive mode for today. Include pending approvals, routine failures, urgent inbox items, and the next meetings. If Gmail or Calendar is not ready, degrade gracefully with setup hints.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 12,456 | 9,074 | -27% | 1 | 1 | 0% | 2,518 | 2,028 | -19% | 0 | 0 | — |
case-02 | fail→fail | 6,181 | 8,219 | +33% | 1 | 1 | 0% | 1,088 | 2,181 | +100% | 0 | 0 | — |
case-03 | fail→pass | 6,599 | 7,620 | +15% | 1 | 1 | 0% | 1,069 | 1,941 | +82% | 0 | 0 | — |
case-04 | pass→pass | 9,245 | 5,838 | -37% | 1 | 1 | 0% | 1,580 | 1,580 | 0% | 0 | 0 | — |
case-05 | pass→pass | 9,318 | 5,849 | -37% | 1 | 1 | 0% | 1,434 | 1,533 | +7% | 0 | 0 | — |
case-06 | pass→fail | 12,605 | 7,524 | -40% | 1 | 1 | 0% | 1,875 | 1,708 | -9% | 0 | 0 | — |
case-07 | pass→pass | 4,102 | 4,542 | +11% | 1 | 1 | 0% | 592 | 1,351 | +128% | 0 | 0 | — |
case-08 | pass→pass | 9,284 | 6,818 | -27% | 1 | 1 | 0% | 1,468 | 1,759 | +20% | 0 | 0 | — |
case-09 | pass→pass | 7,332 | 6,543 | -11% | 1 | 1 | 0% | 1,196 | 1,634 | +37% | 0 | 0 | — |
case-10 | pass→pass | 8,810 | 5,848 | -34% | 1 | 1 | 0% | 1,364 | 1,570 | +15% | 0 | 0 | — |
case-11 | pass→pass | 8,775 | 5,658 | -36% | 1 | 1 | 0% | 1,287 | 1,600 | +24% | 0 | 0 | — |
case-12 | pass→pass | 9,472 | 3,612 | -62% | 1 | 1 | 0% | 1,494 | 1,169 | -22% | 0 | 0 | — |
case-13 | pass→pass | 10,034 | 4,868 | -51% | 1 | 1 | 0% | 1,523 | 1,356 | -11% | 0 | 0 | — |
case-14 | pass→pass | 13,245 | 6,928 | -48% | 1 | 1 | 0% | 2,165 | 1,741 | -20% | 0 | 0 | — |
case-15 | pass→pass | 8,223 | 1,962 | -76% | 1 | 1 | 0% | 1,226 | 912 | -26% | 0 | 0 | — |
case-20 | fail→pass | 8,358 | 7,917 | -5% | 1 | 1 | 0% | 1,324 | 2,049 | +55% | 0 | 0 | — |
case-16 | pass→pass | 11,490 | 8,522 | -26% | 1 | 1 | 0% | 1,828 | 2,104 | +15% | 0 | 0 | — |
case-17 | pass→pass | 9,946 | 3,883 | -61% | 1 | 1 | 0% | 1,545 | 1,259 | -19% | 0 | 0 | — |
case-18 | pass→pass | 12,870 | 6,430 | -50% | 1 | 1 | 0% | 1,971 | 1,625 | -18% | 0 | 0 | — |
case-19 | fail→pass | 9,495 | 3,840 | -60% | 1 | 1 | 0% | 1,437 | 1,189 | -17% | 0 | 0 | — |
case-21 | fail→pass | 11,957 | 3,293 | -72% | 1 | 1 | 0% | 1,905 | 1,054 | -45% | 0 | 0 | — |
case-22 | pass→pass | 13,295 | 8,385 | -37% | 1 | 1 | 0% | 1,968 | 1,868 | -5% | 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. The headline lift of +18 percentage points is the difference between those two pass rates over the 22 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.