Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when pre-flighting Ontario expropriation Forms 1-12 under O.Reg. 363/90 before service or filing — verifying statutory authority citations on Form 2, compensation breakdown completeness on Form 6, 30-day notice math on Form 7, dual-signing-officer compliance, or affidavit-of-service adequacy. Catch defects before they trigger procedural challenges.
.claude/skills/thomasmoreai-forms-1-12-completeness-verification/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 34% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 72% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 156% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 166% | 0% |
Detailed verification requirements for commonly-used expropriation forms.
Checklist:
1. Statutory authority citation:
2. Legal description accuracy:
3. Purpose statement clarity:
4. Service requirements:
5. Dual signing officers (municipal expropriations):
Example defect:
Checklist:
1. Compensation calculation basis:
2. Components breakdown:
3. Payment timing:
4. Acceptance deadline:
Example defect:
Checklist:
1. Minimum 30-day notice period:
2. Compensation offered/paid confirmation:
3. Possession date clarity:
Example defect:
Systematic tracking of frequent errors to prevent recurrence.
Defect: Form cites "Municipal Act" without specific section or subsection.
Example:
Prevention:
Defect: Legal description in Form 2 differs from legal description in registered plan or Form 7.
Example:
Correction:
Prevention:
Defect: No proof of service filed, or incomplete affidavit.
Example:
Complete affidavit should state: > "I, Jane Doe, process server, make oath and say: > 1. On May 1, 2025 at 10:30 AM, I attended at 123 Main Street, Toronto, Ontario. > 2. I personally served John Smith by handing him a copy of Form 2 (Notice of Application for Approval to Expropriate) dated April 25, 2025. > 3. I identified Mr. Smith by asking his name, and he confirmed he is the registered owner of the property. > 4. Attached as Exhibit 'A' is a copy of the Form 2 served. > 5. Sworn before me at Toronto this 1st day of May, 2025."
Prevention:
Defect: Form 6 states total compensation without itemization.
Example:
Correction:
Prevention:
Defect: Only one officer signs municipal expropriation form (municipal by-laws typically require two).
Example:
Correction:
Prevention:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 19,061 | 12,767 | -33% | 1 | 1 | 0% | 3,663 | 4,925 | +34% | 0 | 0 | — |
case-02 | fail→pass | 18,718 | 18,413 | -2% | 1 | 1 | 0% | 3,357 | 5,789 | +72% | 0 | 0 | — |
case-03 | pass→pass | 11,887 | 8,328 | -30% | 1 | 1 | 0% | 2,376 | 4,102 | +73% | 0 | 0 | — |
case-04 | pass→pass | 10,545 | 10,904 | +3% | 1 | 1 | 0% | 1,711 | 4,342 | +154% | 0 | 0 | — |
case-05 | pass→pass | 11,388 | 10,931 | -4% | 1 | 1 | 0% | 1,947 | 4,360 | +124% | 0 | 0 | — |
case-06 | fail→pass | 20,075 | 6,419 | -68% | 1 | 1 | 0% | 3,143 | 3,597 | +14% | 0 | 0 | — |
case-07 | fail→pass | 7,630 | 4,999 | -34% | 1 | 1 | 0% | 1,366 | 3,494 | +156% | 0 | 0 | — |
case-08 | fail→pass | 9,142 | 7,755 | -15% | 1 | 1 | 0% | 1,484 | 3,946 | +166% | 0 | 0 | — |
case-09 | fail→pass | 20,555 | 20,644 | +0% | 1 | 1 | 0% | 3,650 | 5,726 | +57% | 0 | 0 | — |
case-10 | fail→pass | 13,501 | 9,363 | -31% | 1 | 1 | 0% | 1,987 | 4,158 | +109% | 0 | 0 | — |
case-11 | pass→pass | 7,673 | 5,898 | -23% | 1 | 1 | 0% | 1,333 | 3,662 | +175% | 0 | 0 | — |
case-12 | pass→pass | 11,737 | 11,510 | -2% | 1 | 1 | 0% | 1,958 | 4,199 | +114% | 0 | 0 | — |
case-13 | pass→pass | 8,423 | 4,102 | -51% | 1 | 1 | 0% | 1,312 | 3,353 | +156% | 0 | 0 | — |
case-14 | pass→pass | 5,049 | 3,901 | -23% | 1 | 1 | 0% | 741 | 3,305 | +346% | 0 | 0 | — |
case-15 | pass→pass | 12,733 | 7,492 | -41% | 1 | 1 | 0% | 1,802 | 3,689 | +105% | 0 | 0 | — |
case-16 | pass→pass | 8,183 | 8,363 | +2% | 1 | 1 | 0% | 1,591 | 4,040 | +154% | 0 | 0 | — |
case-17 | fail→fail | 12,886 | 9,690 | -25% | 1 | 1 | 0% | 2,238 | 4,696 | +110% | 0 | 0 | — |
case-18 | pass→pass | 12,097 | 11,020 | -9% | 1 | 1 | 0% | 2,123 | 4,733 | +123% | 0 | 0 | — |
case-19 | pass→pass | 13,968 | 10,583 | -24% | 1 | 1 | 0% | 2,292 | 4,375 | +91% | 0 | 0 | — |
case-20 | pass→pass | 19,323 | 23,619 | +22% | 1 | 1 | 0% | 2,662 | 6,509 | +145% | 0 | 0 | — |
case-21 | pass→pass | 23,293 | 33,889 | +45% | 1 | 1 | 0% | 3,596 | 8,086 | +125% | 0 | 0 | — |
case-22 | pass→pass | 19,904 | 22,527 | +13% | 1 | 1 | 0% | 3,274 | 6,280 | +92% | 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 +32 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.