Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Sweep a contracts folder and extract every renewal date, auto-renew clause, notice window, and price-escalation term into a renewal calendar -- then create calendar reminders ahead of each notice deadline via Cowork's calendar write tools. The skill that stops auto-renewals from ambushing you.
.claude/skills/onewave-ai-cowork-contract-renewal-radar/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 22% | 0% |
Read a folder of contracts the way outside counsel prepping a renewal calendar does: find every date-driven obligation, compute the real decision deadline (notice deadline, not renewal date), and put it somewhere it will actually be seen. Complements contract-analyzer (deep single-contract review) -- this skill is the portfolio sweep.
renewal date minus notice period minus a 14-day decision buffer. That is the radar date. Contracts already inside their notice window get flagged URGENT at the top of the report.renewal-calendar.md: chronological table of radar dates for the next 24 months with counterparty, annual value (when stated), what happens if nothing is done, and the verbatim notice clause. Separately list contracts with no end date or missing signature pages as data-quality issues.Renewal decision: [counterparty], description containing the notice deadline, clause quote, and file path -- after showing the user the list of events to be created and getting approval.As a monthly Cowork scheduled task: re-sweep the folder for new/amended contracts, update the calendar, and deliver a digest leading with anything entering its decision window in the next 60 days.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 11,664 | 7,386 | -37% | 1 | 1 | 0% | 2,238 | 1,355 | -39% | 0 | 0 | — |
case-02 | pass→pass | 14,271 | 2,856 | -80% | 1 | 1 | 0% | 2,026 | 1,224 | -40% | 0 | 0 | — |
case-03 | fail→fail | 24,596 | 4,587 | -81% | 1 | 1 | 0% | 4,084 | 1,234 | -70% | 0 | 0 | — |
case-04 | pass→pass | 3,198 | 15,679 | +390% | 1 | 1 | 0% | 444 | 2,880 | +549% | 0 | 0 | — |
case-05 | pass→pass | 10,979 | 13,967 | +27% | 1 | 1 | 0% | 2,060 | 2,815 | +37% | 0 | 0 | — |
case-06 | pass→pass | 20,116 | 20,953 | +4% | 1 | 1 | 0% | 3,048 | 4,033 | +32% | 0 | 0 | — |
case-07 | fail→pass | 10,437 | 7,873 | -25% | 1 | 1 | 0% | 1,671 | 2,023 | +21% | 0 | 0 | — |
case-08 | fail→fail | 8,192 | 5,063 | -38% | 1 | 1 | 0% | 1,320 | 1,592 | +21% | 0 | 0 | — |
case-09 | fail→pass | 14,217 | 8,665 | -39% | 1 | 1 | 0% | 2,045 | 2,239 | +9% | 0 | 0 | — |
case-10 | fail→fail | 8,961 | 10,661 | +19% | 1 | 1 | 0% | 1,432 | 2,388 | +67% | 0 | 0 | — |
case-11 | fail→pass | 10,886 | 9,950 | -9% | 1 | 1 | 0% | 1,718 | 2,286 | +33% | 0 | 0 | — |
case-12 | pass→pass | 9,482 | 4,076 | -57% | 1 | 1 | 0% | 1,651 | 1,359 | -18% | 0 | 0 | — |
case-13 | fail→fail | 9,654 | 2,297 | -76% | 1 | 1 | 0% | 1,487 | 1,084 | -27% | 0 | 0 | — |
case-14 | pass→pass | 11,364 | 8,033 | -29% | 1 | 1 | 0% | 1,764 | 1,933 | +10% | 0 | 0 | — |
case-15 | fail→pass | 8,242 | 5,083 | -38% | 1 | 1 | 0% | 1,321 | 1,529 | +16% | 0 | 0 | — |
case-16 | pass→pass | 10,182 | 3,773 | -63% | 1 | 1 | 0% | 1,527 | 1,212 | -21% | 0 | 0 | — |
case-17 | fail→fail | 15,321 | 15,922 | +4% | 1 | 1 | 0% | 2,443 | 3,166 | +30% | 0 | 0 | — |
case-18 | fail→fail | 6,789 | 4,986 | -27% | 1 | 1 | 0% | 971 | 1,557 | +60% | 0 | 0 | — |
case-19 | fail→pass | 12,891 | 10,280 | -20% | 1 | 1 | 0% | 1,885 | 2,293 | +22% | 0 | 0 | — |
case-20 | fail→pass | 12,671 | 4,798 | -62% | 1 | 1 | 0% | 2,224 | 1,537 | -31% | 0 | 0 | — |
case-21 | pass→pass | 11,725 | 12,639 | +8% | 1 | 1 | 0% | 1,688 | 2,561 | +52% | 0 | 0 | — |
case-22 | pass→pass | 11,122 | 4,592 | -59% | 1 | 1 | 0% | 1,927 | 1,412 | -27% | 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 +27 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.