Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Automate tenant management and lease operations with AppFolio. Trigger: "appfolio tenant management".
.claude/skills/jeremylongshore-appfolio-core-workflow-b/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -35% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-22 | ✗→✓ | ▲ Improved | 14% | 0% |
Manage the full lifecycle of maintenance work orders through AppFolio's Stack API. Use this workflow when tenants submit maintenance requests, when you need to assign vendors to open work orders, track repair progress across properties, or close out completed jobs with cost records. This is the secondary workflow — for property dashboards and leasing, see appfolio-core-workflow-a.
typescriptconst workOrder = await client.workOrders.create({ property_id: 'prop_4821', unit_id: 'unit_12B', category: 'plumbing', priority: 'high', description: 'Kitchen sink leaking under cabinet — tenant reports water damage', requested_by: 'tenant_8934', due_date: '2026-04-10', }); console.log(`Work order ${workOrder.id} created — status: ${workOrder.status}`);
typescriptconst assignment = await client.workOrders.assign(workOrder.id, { vendor_id: 'vendor_plumb_01', scheduled_date: '2026-04-08', time_window: '09:00-12:00', notes: 'Tenant prefers morning. Enter through side gate.', }); console.log(`Assigned to ${assignment.vendor_name} on ${assignment.scheduled_date}`);
typescriptconst open = await client.workOrders.list({ property_id: 'prop_4821', status: ['open', 'in_progress', 'scheduled'], sort: 'priority_desc', }); open.items.forEach(wo => console.log(`#${wo.id} [${wo.priority}] ${wo.category} — ${wo.status} (due ${wo.due_date})`) );
typescriptconst closed = await client.workOrders.close(workOrder.id, { resolution: 'Replaced P-trap and tightened supply line. No further leaks.', labor_cost: 175.00, materials_cost: 42.50, completed_date: '2026-04-08', attachments: ['receipt_plumb_0408.pdf'], }); console.log(`Closed #${closed.id} — total cost: $${closed.total_cost}`);
| Issue | Cause | Fix | |-------|-------|-----| | 401 Unauthorized | Expired or invalid API credentials | Refresh client_id/secret in Stack dashboard | | 404 Work Order Not Found | Wrong work order ID or already deleted | Verify ID with workOrders.list() | | 422 Missing required fields | Category or property_id omitted | Include all required fields per schema | | 409 Conflict | Work order already closed | Check status before attempting close | | 429 Rate Limited | Exceeded 120 requests/minute | Add exponential backoff with 1s base delay |
A successful run creates a work order, assigns it to a vendor with a scheduled service window, and closes it with cost records and resolution notes. The property manager gets a complete audit trail from request through completion.
See appfolio-sdk-patterns for authentication setup and pagination helpers.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-15 | pass→pass | 5,488 | 2,374 | -57% | 1 | 1 | 0% | 872 | 1,344 | +54% | 0 | 0 | — |
case-04 | fail→pass | 7,815 | 3,501 | -55% | 1 | 1 | 0% | 1,667 | 1,583 | -5% | 0 | 0 | — |
case-10 | fail→fail | 6,324 | 2,528 | -60% | 1 | 1 | 0% | 1,122 | 1,250 | +11% | 0 | 0 | — |
case-21 | pass→pass | 7,127 | 3,768 | -47% | 1 | 1 | 0% | 1,116 | 1,445 | +29% | 0 | 0 | — |
case-01 | fail→pass | 11,275 | 5,781 | -49% | 1 | 1 | 0% | 2,454 | 2,108 | -14% | 0 | 0 | — |
case-02 | fail→pass | 13,905 | 8,312 | -40% | 1 | 1 | 0% | 2,849 | 1,839 | -35% | 0 | 0 | — |
case-03 | fail→pass | 10,774 | 4,806 | -55% | 1 | 1 | 0% | 2,133 | 1,842 | -14% | 0 | 0 | — |
case-22 | fail→pass | 14,686 | 5,064 | -66% | 1 | 1 | 0% | 1,523 | 1,741 | +14% | 0 | 0 | — |
case-05 | fail→pass | 11,471 | 3,768 | -67% | 1 | 1 | 0% | 2,387 | 1,743 | -27% | 0 | 0 | — |
case-06 | fail→pass | 19,124 | 4,238 | -78% | 1 | 1 | 0% | 1,665 | 1,616 | -3% | 0 | 0 | — |
case-07 | pass→pass | 9,969 | 4,461 | -55% | 1 | 1 | 0% | 1,832 | 1,862 | +2% | 0 | 0 | — |
case-08 | pass→pass | 11,582 | 5,407 | -53% | 1 | 1 | 0% | 2,163 | 1,985 | -8% | 0 | 0 | — |
case-09 | fail→pass | 10,699 | 4,614 | -57% | 1 | 1 | 0% | 1,746 | 1,672 | -4% | 0 | 0 | — |
case-11 | pass→pass | 9,754 | 4,279 | -56% | 1 | 1 | 0% | 1,636 | 1,610 | -2% | 0 | 0 | — |
case-12 | pass→pass | 11,647 | 3,678 | -68% | 1 | 1 | 0% | 2,143 | 1,640 | -23% | 0 | 0 | — |
case-13 | fail→pass | 4,964 | 1,711 | -66% | 1 | 1 | 0% | 811 | 1,170 | +44% | 0 | 0 | — |
case-14 | fail→pass | 11,986 | 4,623 | -61% | 1 | 1 | 0% | 2,058 | 1,769 | -14% | 0 | 0 | — |
case-16 | fail→pass | 11,018 | 3,581 | -67% | 1 | 1 | 0% | 1,861 | 1,546 | -17% | 0 | 0 | — |
case-17 | fail→pass | 5,919 | 2,080 | -65% | 1 | 1 | 0% | 920 | 1,297 | +41% | 0 | 0 | — |
case-18 | pass→pass | 10,082 | 2,676 | -73% | 1 | 1 | 0% | 1,819 | 1,352 | -26% | 0 | 0 | — |
case-19 | pass→pass | 14,533 | 3,122 | -79% | 1 | 1 | 0% | 2,491 | 1,476 | -41% | 0 | 0 | — |
case-20 | pass→pass | 14,795 | 7,015 | -53% | 1 | 1 | 0% | 2,235 | 1,975 | -12% | 0 | 0 | — |
case-23 | pass→pass | 4,171 | 3,796 | -9% | 1 | 1 | 0% | 742 | 1,553 | +109% | 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. 23 cases were attempted. The headline lift of +52 percentage points is the difference between those two pass rates over the 23 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.