Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Turn a business trip into a one-page brief that runs itself — the itinerary with buffers and failure modes, the meeting logistics pre-solved (addresses, contacts, backup numbers), the packing-and-prep list by trip type, and the expense capture set up before departure. Use when asked prep my business trip, build the travel brief, I always forget something when traveling, or organize this three-city week. Produces the one-page brief: timeline with buffers, the per-meeting logistics, the contingenc
.claude/skills/mohitagw15856-travel-brief/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 81% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 60% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 62% | 0% |
Business trips fail in the seams — the connection that assumed zero delays, the meeting address that lived in an email now unreachable on airport wifi, the receipt pile that becomes March's archaeology (expense-discipline starts before departure). The brief is one page that runs the trip: the timeline with honest buffers (the failure modes priced in), every meeting's logistics pre-solved and offline-accessible, the contingency card (what happens when the flight cancels — decided calmly, not at gate B7), and the capture habits armed before wheels-up.
Ask for these if not provided:
End-to-end with buffers, each named for what it absorbs · timezone math done · transit blocked honestly]
| Meeting | Address (full) | Contact + backup | Arrive by | Prep | |---|---|---|---|---|
Flight cancels → … · meeting moves → … · tech fails → … — each pre-decided with its first phone call]
Brief saved offline · trip expense folder open · receipt habit armed · OOO-lite + delegate · chargers/adapters per the personal failure list]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→pass | 45,459 | 34,621 | -24% | 1 | 1 | 0% | 5,525 | 5,943 | +8% | 0 | 0 | — |
case-03 | fail→fail | 45,414 | 56,784 | +25% | 1 | 1 | 0% | 6,476 | 5,530 | -15% | 0 | 0 | — |
case-01 | fail→pass | 41,263 | 31,925 | -23% | 1 | 1 | 0% | 4,937 | 5,474 | +11% | 0 | 0 | — |
case-04 | fail→pass | 20,528 | 29,533 | +44% | 1 | 1 | 0% | 2,824 | 5,111 | +81% | 0 | 0 | — |
case-05 | fail→pass | 23,514 | 22,660 | -4% | 1 | 1 | 0% | 2,387 | 3,822 | +60% | 0 | 0 | — |
case-06 | fail→fail | 33,343 | 31,262 | -6% | 1 | 1 | 0% | 3,558 | 5,230 | +47% | 0 | 0 | — |
case-07 | fail→fail | 15,172 | 29,462 | +94% | 1 | 1 | 0% | 2,128 | 4,386 | +106% | 0 | 0 | — |
case-08 | fail→pass | 20,928 | 32,639 | +56% | 1 | 1 | 0% | 2,958 | 4,792 | +62% | 0 | 0 | — |
case-09 | fail→pass | 23,201 | 32,074 | +38% | 1 | 1 | 0% | 2,783 | 4,824 | +73% | 0 | 0 | — |
case-10 | fail→pass | 18,681 | 12,018 | -36% | 1 | 1 | 0% | 1,948 | 2,237 | +15% | 0 | 0 | — |
case-11 | fail→fail | 14,878 | 29,174 | +96% | 1 | 1 | 0% | 1,775 | 4,677 | +163% | 0 | 0 | — |
case-12 | fail→fail | 30,858 | 28,223 | -9% | 1 | 1 | 0% | 2,814 | 5,222 | +86% | 0 | 0 | — |
case-13 | fail→pass | 22,882 | 25,522 | +12% | 1 | 1 | 0% | 2,392 | 3,878 | +62% | 0 | 0 | — |
case-14 | fail→fail | 17,221 | 20,485 | +19% | 1 | 1 | 0% | 1,818 | 3,495 | +92% | 0 | 0 | — |
case-15 | fail→pass | 25,193 | 33,483 | +33% | 1 | 1 | 0% | 3,307 | 4,894 | +48% | 0 | 0 | — |
case-16 | fail→pass | 23,148 | 23,242 | +0% | 1 | 1 | 0% | 2,553 | 3,768 | +48% | 0 | 0 | — |
case-17 | fail→pass | 27,194 | 30,158 | +11% | 1 | 1 | 0% | 3,419 | 5,115 | +50% | 0 | 0 | — |
case-18 | pass→pass | 14,612 | 24,972 | +71% | 1 | 1 | 0% | 2,185 | 3,979 | +82% | 0 | 0 | — |
case-19 | fail→pass | 14,202 | 22,561 | +59% | 1 | 1 | 0% | 2,000 | 3,663 | +83% | 0 | 0 | — |
case-20 | pass→pass | 21,699 | 22,032 | +2% | 1 | 1 | 0% | 2,393 | 4,086 | +71% | 0 | 0 | — |
case-21 | pass→pass | 15,519 | 22,043 | +42% | 1 | 1 | 0% | 2,439 | 3,618 | +48% | 0 | 0 | — |
case-22 | pass→pass | 25,558 | 41,324 | +62% | 1 | 1 | 0% | 3,416 | 5,301 | +55% | 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 +55 percentage points is the difference between those two pass rates over the 22 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.