Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Run a company event — the launch party, the customer day, the team celebration — as the operation it is: the goal that shapes every choice, the budget with its forgotten lines, the vendor and venue coordination, the run-of-show with owners, and the day-of roles that keep hosts hosting. Use when asked plan the company event, organize our customer day/holiday party/launch event, what am I forgetting for this event, or be the run-of-show for Thursday. Produces the goal-shaped plan, the budget with
.claude/skills/mohitagw15856-company-event-ops/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 44% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 45% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 74% | 0% |
Company events are wedding-logistics-planner mechanics pointed at a business goal — and the goal is the part that usually goes unstated, which is why so many events are pleasant and pointless. "What is this event for" (celebrate the team? deepen customer relationships? mark the launch loudly?) shapes every downstream choice: venue, format, guest list, budget, and what success means Friday morning. Below the goal, it's operations: the budget with its routinely-forgotten lines (AV, service fees, the buffer), vendors coordinated on one sheet, the run-of-show with a named owner per segment, and the day-of rule that saves every event: the hosts host; the runners run — decided in advance, or the CEO spends the customer day moving chairs.
Ask for these if not provided:
The stated goal · the choices it drove: venue/format/program/guest-list — each with its why]
The lines including AV, service+gratuity, signage, overtime clauses · the 10% buffer · per-head]
Vendor cards: contact/arrival/needs/payment · The minute-map: time × segment × owner × contingency]
Hosts (protected) · runners (+ their budget cap) · decision-holder · vendor-point — names against all]
The follow-through: thank-yous, the share, the autopsy note — owned and dated]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-07 | pass→pass | 13,293 | 19,487 | +47% | 1 | 1 | 0% | 2,463 | 4,812 | +95% | 0 | 0 | — |
case-01 | fail→fail | 31,288 | 18,941 | -39% | 1 | 1 | 0% | 5,440 | 4,919 | -10% | 0 | 0 | — |
case-02 | fail→fail | 25,005 | 18,970 | -24% | 1 | 1 | 0% | 4,436 | 4,750 | +7% | 0 | 0 | — |
case-03 | fail→pass | 28,290 | 21,521 | -24% | 1 | 1 | 0% | 4,953 | 5,377 | +9% | 0 | 0 | — |
case-04 | fail→pass | 14,592 | 14,296 | -2% | 1 | 1 | 0% | 2,896 | 4,156 | +44% | 0 | 0 | — |
case-05 | fail→pass | 17,886 | 14,344 | -20% | 1 | 1 | 0% | 2,812 | 3,605 | +28% | 0 | 0 | — |
case-06 | pass→pass | 20,993 | 24,856 | +18% | 1 | 1 | 0% | 3,433 | 5,419 | +58% | 0 | 0 | — |
case-08 | fail→pass | 17,163 | 17,539 | +2% | 1 | 1 | 0% | 2,990 | 4,331 | +45% | 0 | 0 | — |
case-09 | pass→pass | 13,921 | 14,550 | +5% | 1 | 1 | 0% | 2,165 | 3,713 | +72% | 0 | 0 | — |
case-10 | pass→pass | 12,710 | 14,514 | +14% | 1 | 1 | 0% | 2,323 | 3,764 | +62% | 0 | 0 | — |
case-11 | fail→pass | 15,266 | 20,053 | +31% | 1 | 1 | 0% | 2,746 | 4,768 | +74% | 0 | 0 | — |
case-12 | fail→pass | 12,780 | 13,142 | +3% | 1 | 1 | 0% | 2,315 | 3,715 | +60% | 0 | 0 | — |
case-13 | pass→pass | 12,748 | 18,247 | +43% | 1 | 1 | 0% | 2,486 | 4,181 | +68% | 0 | 0 | — |
case-14 | pass→pass | 15,382 | 16,873 | +10% | 1 | 1 | 0% | 2,458 | 3,945 | +60% | 0 | 0 | — |
case-15 | fail→pass | 14,383 | 13,120 | -9% | 1 | 1 | 0% | 2,306 | 3,442 | +49% | 0 | 0 | — |
case-16 | pass→pass | 17,204 | 20,034 | +16% | 1 | 1 | 0% | 2,625 | 4,499 | +71% | 0 | 0 | — |
case-17 | fail→pass | 11,987 | 17,537 | +46% | 1 | 1 | 0% | 1,972 | 3,881 | +97% | 0 | 0 | — |
case-18 | pass→pass | 11,767 | 10,583 | -10% | 1 | 1 | 0% | 1,877 | 2,958 | +58% | 0 | 0 | — |
case-19 | fail→fail | 15,113 | 12,581 | -17% | 1 | 1 | 0% | 2,531 | 3,198 | +26% | 0 | 0 | — |
case-20 | fail→pass | 10,219 | 8,048 | -21% | 1 | 1 | 0% | 1,753 | 2,675 | +53% | 0 | 0 | — |
case-21 | fail→fail | 20,675 | 26,731 | +29% | 1 | 1 | 0% | 3,623 | 6,042 | +67% | 0 | 0 | — |
case-22 | fail→fail | 12,824 | 15,007 | +17% | 1 | 1 | 0% | 2,218 | 3,770 | +70% | 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 +41 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.