@wonsukchoi/crondex 0.22.0 → 0.23.0

package/jobs/retail/price-tag-audit-accuracy-check.yaml

@@ -0,0 +1,37 @@
+id: price-tag-audit-accuracy-check
+version: 1
+name: Price Tag Audit Accuracy Check
+description: >
+  Reviews photos of shelf price tags against the current POS price file
+  and flags mismatches. Use this if a customer has ever caught a
+  scan-versus-tag price discrepancy at the register before the store did
+  — those add up fast in refunds, complaints, and occasional fines.
+category: retail
+tags: [retail, pricing, audit, compliance]
+schedule: "0 8 * * 4"
+timezone: "America/New_York"
+runner: agent-prompt
+prompt: |
+  Compare this week's shelf tag photos ({{tag_photos_path}}) against the
+  current POS price export ({{pos_price_file}}) for store {{store_id}}.
+  1. Read the SKU and posted price off each tag photo (OCR/read the
+     image directly).
+  2. Match each tag to its POS price by SKU and flag any mismatch,
+     noting whether the tag is higher or lower than the register price.
+  3. Prioritize mismatches where the tag price is lower than the POS
+     price — those are the ones customers will notice and complain about
+     first, and in some states are a pricing-accuracy compliance risk.
+  4. Summarize as a punch list: SKU, tag price, POS price, and which tag
+     needs reprinting.
+variables:
+  tag_photos_path:
+    default: "./shelf-tag-photos/"
+    description: Folder of this week's shelf price tag photos for the store.
+  pos_price_file:
+    default: "./pos-price-export.csv"
+    description: Current POS price export, columns sku,price.
+  store_id:
+    default: "store-001"
+    description: Identifier for the store being audited.
+compatible_agents: [claude, codex, hermes, openclaw, generic]
+notes: Reading prices off photographed tags and matching them against a price list needs visual judgment a fixed script can't reliably do.

package/jobs/retail/sales-per-square-foot-tracker.yaml

@@ -0,0 +1,36 @@
+id: sales-per-square-foot-tracker
+version: 1
+name: Sales Per Square Foot Tracker
+description: >
+  Calculates each store's sales-per-square-foot for the period and flags
+  stores falling below a benchmark. Use this if underperforming store
+  footprints have ever gone unnoticed until an annual real-estate review
+  instead of getting flagged while there was still time to act.
+category: retail
+tags: [retail, sales, performance, benchmarking]
+schedule: "0 7 1 * *"
+timezone: "UTC"
+runner: shell
+command: >
+  awk -F',' -v floor="{{benchmark_sales_per_sqft}}" '
+    NR==1 {next}
+    {
+      sqft = $3;
+      if (sqft > 0) {
+        rate = $2 / sqft;
+        status = (rate < floor) ? "BELOW BENCHMARK" : "ok";
+        if (status == "BELOW BENCHMARK") {
+          printf "BELOW BENCHMARK: store %s — $%.2f/sqft (benchmark $%.2f/sqft)\n", $1, rate, floor;
+        }
+      }
+    }
+  ' "{{store_sales_csv}}"
+variables:
+  store_sales_csv:
+    default: "./store-sales-sqft.csv"
+    description: "CSV with header row, columns: store_id,period_sales_usd,selling_sqft — one row per store for the period."
+  benchmark_sales_per_sqft:
+    default: 300
+    description: Minimum acceptable sales-per-square-foot for the period before a store gets flagged.
+compatible_agents: [generic, claude, codex]
+notes: A single low period isn't necessarily a problem store — track the trend across periods before drawing conclusions about a location.

package/jobs/retail/seasonal-staffing-coverage-check.yaml

@@ -0,0 +1,35 @@
+id: seasonal-staffing-coverage-check
+version: 1
+name: Seasonal Staffing Coverage Check
+description: >
+  Compares scheduled labor hours against a forecasted seasonal-peak
+  staffing plan and flags stores that are under-scheduled. Use this if a
+  store has ever gone into a holiday or back-to-school rush understaffed
+  because the seasonal plan and the actual schedule quietly drifted apart.
+category: retail
+tags: [retail, staffing, seasonal, scheduling]
+schedule: "0 9 * * 1"
+timezone: "America/New_York"
+runner: shell
+command: >
+  awk -F',' -v gap_pct="{{coverage_gap_threshold_pct}}" '
+    NR==1 {next}
+    {
+      planned = $2; scheduled = $3;
+      if (planned > 0) {
+        gap = ((planned - scheduled) / planned) * 100;
+        if (gap >= gap_pct) {
+          printf "UNDER-SCHEDULED: store %s — planned %.0f hrs, scheduled %.0f hrs (%.1f%% short)\n", $1, planned, scheduled, gap;
+        }
+      }
+    }
+  ' "{{seasonal_staffing_csv}}"
+variables:
+  seasonal_staffing_csv:
+    default: "./seasonal-staffing-plan.csv"
+    description: "CSV with header row, columns: store_id,planned_hours,scheduled_hours — one row per store for the upcoming seasonal window."
+  coverage_gap_threshold_pct:
+    default: 10
+    description: Percentage shortfall between planned and scheduled hours that triggers a flag.
+compatible_agents: [generic, claude, codex]
+notes: Complements `employee-shift-gap-check` (day-to-day unfilled shifts) by checking whether the whole store is resourced against a seasonal peak plan, weeks ahead of the rush rather than shift by shift.

package/jobs/retail/store-closing-checklist-compliance-check.yaml

@@ -0,0 +1,32 @@
+id: store-closing-checklist-compliance-check
+version: 1
+name: Store Closing Checklist Compliance Check
+description: >
+  Reviews logged closing checklist completions and flags stores or shifts
+  that skipped required steps, rather than just reminding staff of the
+  checklist. Use this if a store has ever opened the next morning to a
+  safe left unlocked, alarm not set, or cash not dropped because closing
+  execution was never actually checked afterward.
+category: retail
+tags: [retail, checklist, closing, compliance, loss-prevention]
+schedule: "0 7 * * *"
+timezone: "America/New_York"
+runner: shell
+command: >
+  awk -F',' '
+    NR==1 {next}
+    {
+      if ($3 == "incomplete" || $3 == "") {
+        printf "INCOMPLETE CLOSE: store %s on %s — missing step: %s\n", $1, $2, $4;
+      }
+    }
+  ' "{{closing_log_csv}}"
+variables:
+  closing_log_csv:
+    default: "./closing-checklist-log.csv"
+    description: "CSV with header row, columns: store_id,close_date,status(complete/incomplete),missing_step — one row per required closing step per night."
+compatible_agents: [generic, claude, codex]
+notes: >
+  Distinct from `store-opening-checklist-reminder`, which just prints the
+  standard checklist as a reminder — this audits a logged closing record
+  after the fact to catch steps that were actually skipped.

package/jobs/retail/visual-merchandising-compliance-check.yaml

@@ -0,0 +1,40 @@
+id: visual-merchandising-compliance-check
+version: 1
+name: Visual Merchandising Compliance Check
+description: >
+  Reviews store-front and window display photos against the current
+  visual merchandising directive and flags stores that haven't executed
+  it. Use this if a seasonal window or entrance display has ever stayed
+  up stale for weeks after corporate sent the new directive.
+category: retail
+tags: [retail, merchandising, visual, compliance]
+schedule: "0 8 * * 5"
+timezone: "America/New_York"
+runner: agent-prompt
+prompt: |
+  Compare this week's store-front/window/entrance photos
+  ({{store_photos_path}}) against the current visual merchandising
+  directive ({{vm_directive_path}}) for store {{store_id}}.
+  1. Note whether the required theme, signage, mannequin styling, and
+     entrance table match the directive.
+  2. Flag anything stale (a prior season's display still up) separately
+     from anything simply missing.
+  3. Note any obviously damaged or faded display elements that need
+     replacing regardless of the current directive.
+  4. Give a short, actionable punch list, not a general aesthetic
+     critique.
+variables:
+  store_photos_path:
+    default: "./vm-photos/"
+    description: Folder of this week's store-front and window display photos.
+  vm_directive_path:
+    default: "./current-vm-directive.pdf"
+    description: Path to the current corporate visual merchandising directive.
+  store_id:
+    default: "store-001"
+    description: Identifier for the store being reviewed.
+compatible_agents: [claude, codex, hermes, openclaw, generic]
+notes: >
+  Distinct from `planogram-compliance-audit-check`, which audits shelf/bay
+  product placement inside the store — this covers window, entrance, and
+  seasonal display execution instead.

package/jobs/veterinary/appointment-no-show-rate-tracker.yaml

@@ -0,0 +1,49 @@
+id: appointment-no-show-rate-tracker
+version: 1
+name: Appointment No-Show Rate Tracker
+description: >
+  Calculates the weekly no-show rate from a logged appointment history and
+  flags when it crosses a threshold. Use this if the clinic has ever kept
+  eating empty exam-room slots without noticing the no-show rate had
+  crept up until the schedule was visibly hurting revenue — this tracks
+  the aggregate trend, distinct from a per-appointment outreach job.
+category: veterinary
+tags: [veterinary, appointments, no-show, scheduling, metrics]
+schedule: "0 7 * * 1"
+timezone: "UTC"
+runner: shell
+command: >
+  awk -F',' -v cutoff="{{lookback_days}}" -v threshold="{{alert_threshold_pct}}" '
+    NR==1 {next}
+    {
+      cmd = "date -d \"" $2 "\" +%s 2>/dev/null || date -j -f %Y-%m-%d \"" $2 "\" +%s 2>/dev/null";
+      cmd | getline appt_epoch; close(cmd);
+      "date +%s" | getline now_epoch; close("date +%s");
+      age_days = (now_epoch - appt_epoch) / 86400;
+      if (age_days >= 0 && age_days <= cutoff) {
+        total++;
+        if ($3 == "no-show") noshow++;
+      }
+    }
+    END {
+      if (total > 0) {
+        rate = (noshow / total) * 100;
+        printf "NO-SHOW RATE: %d/%d appointments in last %d day(s) = %.1f%%\n", noshow, total, cutoff, rate;
+        if (rate >= threshold) printf "ALERT: no-show rate %.1f%% is at or above the %.1f%% threshold\n", rate, threshold;
+      } else {
+        print "No appointments found in the lookback window.";
+      }
+    }
+  ' "{{appointments_csv_path}}"
+variables:
+  appointments_csv_path:
+    default: "./appointments.csv"
+    description: "Path to a CSV with columns: pet_name,appointment_date (YYYY-MM-DD),status (arrived/no-show/cancelled)."
+  lookback_days:
+    default: 7
+    description: How many trailing days of appointments to include in the rate calculation.
+  alert_threshold_pct:
+    default: 10
+    description: No-show rate percentage that triggers an alert.
+compatible_agents: [generic, claude, codex]
+notes: Zero-token trend check — pair with a per-appointment outreach job if you also want to contact each no-show individually.

package/jobs/veterinary/boarding-kennel-capacity-utilization-check.yaml

@@ -0,0 +1,39 @@
+id: boarding-kennel-capacity-utilization-check
+version: 1
+name: Boarding/Kennel Capacity Utilization Check
+description: >
+  Compares today's occupied boarding runs/kennels against total capacity
+  and warns when utilization crosses a threshold. Use this if the front
+  desk has ever accepted a boarding reservation only to discover at
+  drop-off that every run was already full — this tracks overall
+  occupancy, distinct from boarding-checkout-reminder.yaml which tracks
+  individual checkout dates.
+category: veterinary
+tags: [veterinary, boarding, kennel, capacity, occupancy]
+schedule: "0 6 * * *"
+timezone: "UTC"
+runner: shell
+command: >
+  today=$(date +%Y-%m-%d);
+  occupied=$(awk -F',' -v today="$today" '
+    NR==1 {next}
+    $3 <= today && $4 >= today {count++}
+    END {print count+0}
+  ' "{{boarding_csv_path}}");
+  total="{{total_kennel_capacity}}";
+  warn="{{warn_threshold_pct}}";
+  pct=$(awk -v o="$occupied" -v t="$total" 'BEGIN { printf "%.1f", (t>0)?(o/t*100):0 }');
+  echo "OCCUPANCY: $occupied/$total runs occupied today ($pct%)";
+  awk -v pct="$pct" -v warn="$warn" 'BEGIN { if (pct+0 >= warn+0) print "ALERT: occupancy at or above " warn "% — confirm capacity before accepting more reservations." }'
+variables:
+  boarding_csv_path:
+    default: "./boarding-schedule.csv"
+    description: "Path to a CSV with columns: pet_name,owner_name,checkin_date (YYYY-MM-DD),checkout_date (YYYY-MM-DD)."
+  total_kennel_capacity:
+    default: 20
+    description: Total number of boarding runs/kennels available.
+  warn_threshold_pct:
+    default: 90
+    description: Occupancy percentage that triggers a capacity warning.
+compatible_agents: [generic, claude, codex]
+notes: Zero-token. Counts any stay whose check-in/checkout window spans today — adjust the CSV export cadence to keep this current through the day.

package/jobs/veterinary/client-payment-plan-delinquency-check.yaml

@@ -0,0 +1,36 @@
+id: client-payment-plan-delinquency-check
+version: 1
+name: Client Payment Plan Delinquency Check
+description: >
+  Checks active client payment plans (e.g. for a large surgery or
+  emergency bill) against their installment schedule and flags anyone
+  who's missed a payment. Use this if a payment plan has ever quietly
+  gone delinquent for months because nobody was cross-checking installment
+  due dates against what actually came in.
+category: veterinary
+tags: [veterinary, billing, payment-plans, collections, clients]
+schedule: "0 9 1,15 * *"
+timezone: "UTC"
+runner: shell
+command: >
+  today=$(date +%s);
+  awk -F',' -v today="$today" -v grace_days="{{grace_period_days}}" '
+    NR==1 {next}
+    $4 == "unpaid" {
+      cmd = "date -d \"" $3 "\" +%s 2>/dev/null || date -j -f %Y-%m-%d \"" $3 "\" +%s 2>/dev/null";
+      cmd | getline due_epoch; close(cmd);
+      overdue_days = (today - due_epoch) / 86400;
+      if (overdue_days >= grace_days) {
+        printf "DELINQUENT: %s — installment $%s due %s (%d days overdue), plan total $%s\n", $1, $2, $3, overdue_days, $5;
+      }
+    }
+  ' "{{payment_plans_csv_path}}"
+variables:
+  payment_plans_csv_path:
+    default: "./payment-plans.csv"
+    description: "Path to a CSV with columns: owner_name,installment_amount,due_date (YYYY-MM-DD),status (paid/unpaid),plan_total_amount."
+  grace_period_days:
+    default: 5
+    description: Days past the due date before a missed installment is flagged as delinquent.
+compatible_agents: [generic, claude, codex]
+notes: Read-only aging check — actual collections outreach and any plan renegotiation should follow your practice's billing policy.

package/jobs/veterinary/dvm-license-dea-registration-expiry-check.yaml

@@ -0,0 +1,41 @@
+id: dvm-license-dea-registration-expiry-check
+version: 1
+name: Veterinarian License & DEA Registration Expiry Check
+description: >
+  Checks each veterinarian's state license and DEA registration expiry
+  dates and warns before either lapses. Use this if a vet has ever
+  discovered a lapsed DEA registration only when trying to write a
+  controlled-substance prescription — distinct from
+  license-credential-expiry-check.yaml in healthcare, which tracks
+  general staff credentials rather than the DVM license plus DEA
+  dual-registration requirement specific to prescribing controlled
+  substances for animal patients.
+category: veterinary
+tags: [veterinary, compliance, licensing, dea, staff]
+schedule: "0 8 * * 1"
+timezone: "UTC"
+runner: shell
+command: >
+  awk -F',' -v days="{{warn_days_before}}" '
+    NR==1 {next}
+    {
+      cmd = "date -d \"" $3 "\" +%s 2>/dev/null || date -j -f %Y-%m-%d \"" $3 "\" +%s 2>/dev/null";
+      cmd | getline license_expiry_epoch; close(cmd);
+      cmd2 = "date -d \"" $4 "\" +%s 2>/dev/null || date -j -f %Y-%m-%d \"" $4 "\" +%s 2>/dev/null";
+      cmd2 | getline dea_expiry_epoch; close(cmd2);
+      "date +%s" | getline now_epoch; close("date +%s");
+      license_days = (license_expiry_epoch - now_epoch) / 86400;
+      dea_days = (dea_expiry_epoch - now_epoch) / 86400;
+      if (license_days <= days) printf "EXPIRING: %s — state license expires %s (%d days)\n", $1, $3, license_days;
+      if (dea_days <= days) printf "EXPIRING: %s — DEA registration expires %s (%d days)\n", $1, $4, dea_days;
+    }
+  ' "{{credential_csv_path}}"
+variables:
+  credential_csv_path:
+    default: "./dvm-credentials.csv"
+    description: "Path to a CSV with columns: veterinarian_name,state,license_expiry_date (YYYY-MM-DD),dea_expiry_date (YYYY-MM-DD)."
+  warn_days_before:
+    default: 60
+    description: How many days before expiry to start warning.
+compatible_agents: [generic, claude, codex]
+notes: Assumes a maintained CSV export from your credentialing/HR system. A lapsed DEA registration blocks legal prescribing of controlled substances — treat DEA warnings as higher priority than license warnings.

package/jobs/veterinary/microchip-registration-verification-check.yaml

@@ -0,0 +1,40 @@
+id: microchip-registration-verification-check
+version: 1
+name: Microchip Registration Verification Check
+description: >
+  Reviews recently implanted microchips and checks whether the chip was
+  actually registered with the manufacturer's database under the owner's
+  current contact info. Use this if a lost pet has ever been scanned at
+  a shelter, chip found, but the registry came back with no owner info
+  or an outdated phone number — the chip is only useful if the
+  registration behind it is current.
+category: veterinary
+tags: [veterinary, microchip, registration, patients, compliance]
+schedule: "0 9 * * 1"
+timezone: "UTC"
+runner: agent-prompt
+prompt: |
+  Review the list of pets microchipped in the last {{lookback_days}} days
+  from {{practice_management_system}}: {{unregistered_chips_list}}.
+  1. For each chip, note whether the record shows it was registered with
+     the manufacturer's registry (e.g. HomeAgain, 24PetWatch, AKC
+     Reunite) and, if so, whether the owner contact info on file matches
+     the registry.
+  2. List any chip implanted but with no registration confirmed, and any
+     chip where the owner's phone/email on file has since changed
+     without the registry being updated.
+  3. Draft a short reminder message for owners with unregistered or
+     stale-registration chips, explaining briefly why an unregistered
+     chip won't reunite them with a lost pet.
+variables:
+  practice_management_system:
+    default: "your practice management system"
+    description: Name of the system where microchip implant records live.
+  unregistered_chips_list:
+    default: "the clinic's microchip implant log for the period"
+    description: Source of implanted-chip records to check against registry status.
+  lookback_days:
+    default: 30
+    description: How far back to check for recently implanted chips.
+compatible_agents: [claude, codex, hermes, openclaw, generic]
+notes: Draft-only outreach — registry lookups may require checking the specific manufacturer's site or portal, since there's no single universal microchip registry.

package/jobs/veterinary/patient-record-completeness-audit.yaml

@@ -0,0 +1,41 @@
+id: patient-record-completeness-audit
+version: 1
+name: Patient (Pet) Record Completeness Audit
+description: >
+  Spot-checks a sample of recent patient records for missing required
+  fields — weight, vaccine history, allergy flags, microchip number — and
+  lists the gaps. Use this if a chart has ever been missing a documented
+  drug allergy right when it mattered, because record completeness was
+  never actually reviewed, only assumed.
+category: veterinary
+tags: [veterinary, records, compliance, audit, patients]
+schedule: "0 10 * * 3"
+timezone: "UTC"
+runner: agent-prompt
+prompt: |
+  Review a sample of {{sample_size}} recent patient records from
+  {{practice_management_system}}, prioritizing patients seen in the last
+  {{lookback_days}} days.
+  1. For each record, check whether the following are present and current:
+     current weight, vaccine history, known allergies (or "none recorded"
+     explicitly, not just blank), and microchip number.
+  2. List each record with missing fields, grouped by which field is
+     missing most often across the sample — that's usually the intake
+     step actually breaking down.
+  3. Flag any record where an allergy field is blank rather than
+     explicitly marked "none" — a blank field is ambiguous and riskier
+     than a marked "none."
+  4. Suggest one process fix if the same field is missing across most of
+     the sample (e.g. "weight isn't being logged at check-in").
+variables:
+  practice_management_system:
+    default: "your practice management system"
+    description: Name of the system where patient records live.
+  sample_size:
+    default: 20
+    description: Number of recent records to sample for the audit.
+  lookback_days:
+    default: 30
+    description: How far back to pull records from for the sample.
+compatible_agents: [claude, codex, hermes, openclaw, generic]
+notes: Sampling audit, not a full chart review — use findings to spot a broken intake step, not to grade individual staff.

package/jobs/veterinary/pet-medication-refill-backlog-check.yaml

@@ -0,0 +1,36 @@
+id: pet-medication-refill-backlog-check
+version: 1
+name: Pet Medication Refill Request Backlog Check
+description: >
+  Checks pending prescription/food refill requests against how long
+  they've been waiting and flags anything past a reasonable turnaround.
+  Use this if a client has ever run out of a pet's chronic medication
+  (e.g. thyroid or seizure medication) because a refill request sat
+  unanswered in the queue over a weekend.
+category: veterinary
+tags: [veterinary, prescriptions, refills, patients, medication]
+schedule: "0 9,15 * * 1-5"
+timezone: "UTC"
+runner: shell
+command: >
+  today=$(date +%s);
+  awk -F',' -v today="$today" -v hours="{{escalate_after_hours}}" '
+    NR==1 {next}
+    $4 == "pending" {
+      cmd = "date -d \"" $3 "\" +%s 2>/dev/null || date -j -f %Y-%m-%d \"" $3 "\" +%s 2>/dev/null";
+      cmd | getline requested_epoch; close(cmd);
+      age_hours = (today - requested_epoch) / 3600;
+      if (age_hours >= hours) {
+        printf "REFILL PENDING: %s (%s) — %s requested %s (%d hrs ago)\n", $1, $2, $5, $3, age_hours;
+      }
+    }
+  ' "{{refill_requests_csv_path}}"
+variables:
+  refill_requests_csv_path:
+    default: "./pet-refill-requests.csv"
+    description: "Path to a CSV with columns: pet_name,owner_name,requested_date (YYYY-MM-DD),status (pending/approved/denied),medication."
+  escalate_after_hours:
+    default: 48
+    description: Hours a refill request can sit pending before it's flagged.
+compatible_agents: [generic, claude, codex]
+notes: Read-only triage — actual approval of a refill is a clinical decision made by the veterinarian, not this job. Flag chronic/maintenance medications for priority handling.

package/jobs/veterinary/puppy-kitten-vaccine-series-compliance-check.yaml

@@ -0,0 +1,39 @@
+id: puppy-kitten-vaccine-series-compliance-check
+version: 1
+name: Puppy/Kitten Vaccine Series Compliance Check
+description: >
+  Tracks young patients partway through their initial multi-dose vaccine
+  series and flags anyone who missed the next dose window. Use this if a
+  puppy or kitten has ever fallen off after the first shot in a series,
+  leaving them under-protected because nobody tracked series progress
+  separately from routine adult booster due dates — distinct from
+  pet-vaccine-due-reminder.yaml, which tracks single due-date reminders
+  rather than a multi-dose series with its own interval rules.
+category: veterinary
+tags: [veterinary, vaccine, puppy, kitten, patients]
+schedule: "0 8 * * *"
+timezone: "UTC"
+runner: agent-prompt
+prompt: |
+  Review the puppy/kitten vaccine series tracking export from
+  {{practice_management_system}} for patients currently mid-series
+  (i.e. have had at least one dose of a multi-dose series but not the
+  final dose).
+  1. For each patient, note the series (e.g. DHPP, FVRCP), which dose
+     number they're on, the date of their last dose, and the
+     recommended interval to the next dose ({{series_interval_weeks}}
+     weeks is typical but confirm against the specific vaccine's label).
+  2. Flag anyone past their expected next-dose window with no
+     appointment booked.
+  3. Sort by most overdue first, since a gap in a puppy/kitten series
+     can mean restarting the series depending on how long the gap was.
+  4. Draft a short reminder message per overdue patient for the owner.
+variables:
+  practice_management_system:
+    default: "your practice management system"
+    description: Name of the system where vaccine series records live.
+  series_interval_weeks:
+    default: 3
+    description: Typical interval in weeks between doses in a puppy/kitten vaccine series (varies by vaccine and product label).
+compatible_agents: [claude, codex, hermes, openclaw, generic]
+notes: Draft-only — always confirm dosing interval and restart rules against the specific vaccine manufacturer's label, since gap tolerance varies by product.

package/jobs/veterinary/senior-pet-wellness-recall-tracker.yaml

@@ -0,0 +1,43 @@
+id: senior-pet-wellness-recall-tracker
+version: 1
+name: Senior Pet Wellness Recall Tracker
+description: >
+  Finds senior pets (past a configurable age threshold) who are overdue
+  for a semi-annual senior wellness exam or bloodwork panel and drafts
+  outreach. Use this if an older pet has ever gone a full year between
+  visits when twice-yearly senior screening would have caught a treatable
+  condition early — distinct from pet-vaccine-due-reminder.yaml, which
+  tracks vaccine due dates rather than age-based wellness screening
+  cadence.
+category: veterinary
+tags: [veterinary, senior-pets, wellness, recall, outreach]
+schedule: "0 8 1 * *"
+timezone: "UTC"
+runner: agent-prompt
+prompt: |
+  Check {{practice_management_system}} for patients aged
+  {{senior_age_threshold_years}} years or older whose last wellness
+  exam or senior bloodwork panel was more than
+  {{recall_interval_months}} months ago, with no future appointment
+  booked.
+  1. List each pet, species/breed if known, last visit date, and how
+     overdue they are.
+  2. Sort by most overdue first.
+  3. Draft a short outreach message per pet (or a reusable template)
+     inviting the owner to book a senior wellness visit, mentioning
+     that early screening catches age-related conditions while they're
+     still manageable.
+  4. Separately call out any pet more than double the recall interval
+     overdue.
+variables:
+  practice_management_system:
+    default: "your practice management system"
+    description: Name of the system where appointment history lives.
+  senior_age_threshold_years:
+    default: 7
+    description: Age in years at which a pet is treated as a senior for recall purposes.
+  recall_interval_months:
+    default: 6
+    description: How many months between senior wellness visits before a pet counts as due.
+compatible_agents: [claude, codex, hermes, openclaw, generic]
+notes: Draft-only — outreach messages should be reviewed by staff before sending. Senior age threshold varies by species/breed; adjust the default per your patient mix.

package/jobs/veterinary/spay-neuter-followup-compliance-check.yaml

@@ -0,0 +1,35 @@
+id: spay-neuter-followup-compliance-check
+version: 1
+name: Spay/Neuter Post-Op Follow-Up Compliance Check
+description: >
+  Checks which spay/neuter patients are due for their post-op recheck
+  (incision check, suture removal) and haven't had one logged. Use this
+  if a pet has ever developed an incision complication that would have
+  been caught at a recheck nobody scheduled or confirmed happened.
+category: veterinary
+tags: [veterinary, surgery, post-op, follow-up, compliance]
+schedule: "0 8 * * *"
+timezone: "UTC"
+runner: shell
+command: >
+  today=$(date +%s);
+  awk -F',' -v today="$today" -v window="{{recheck_window_days}}" '
+    NR==1 {next}
+    $4 != "completed" {
+      cmd = "date -d \"" $2 "\" +%s 2>/dev/null || date -j -f %Y-%m-%d \"" $2 "\" +%s 2>/dev/null";
+      cmd | getline surgery_epoch; close(cmd);
+      days_since = (today - surgery_epoch) / 86400;
+      if (days_since >= window) {
+        printf "NO RECHECK LOGGED: %s (%s) — surgery %s, %d days ago, recheck status: %s\n", $1, $3, $2, days_since, $4;
+      }
+    }
+  ' "{{spay_neuter_csv_path}}"
+variables:
+  spay_neuter_csv_path:
+    default: "./spay-neuter-cases.csv"
+    description: "Path to a CSV with columns: pet_name,surgery_date (YYYY-MM-DD),owner_name,recheck_status (scheduled/completed/none)."
+  recheck_window_days:
+    default: 10
+    description: Days after surgery by which a post-op recheck should be scheduled or completed.
+compatible_agents: [generic, claude, codex]
+notes: Assumes a maintained export from your practice management system — this is a backstop for cases that fell out of the recheck workflow, not a substitute for discharge instructions.

package/jobs/veterinary/specialist-referral-followup-check.yaml

@@ -0,0 +1,38 @@
+id: specialist-referral-followup-check
+version: 1
+name: Veterinary Specialist Referral Follow-Up Check
+description: >
+  Checks outbound referrals to veterinary specialists (cardiology,
+  oncology, orthopedic surgery, etc.) that haven't been scheduled or
+  completed within a set window and flags them for follow-up. Use this
+  if a pet has ever been referred out for a serious condition and the
+  primary clinic never confirmed the owner actually followed through and
+  got the specialist appointment.
+category: veterinary
+tags: [veterinary, referrals, specialists, follow-up, care-coordination]
+schedule: "0 8 * * 1-5"
+timezone: "UTC"
+runner: shell
+command: >
+  today=$(date +%s);
+  awk -F',' -v today="$today" -v days="{{stale_days}}" '
+    NR==1 {next}
+    $4 != "" {next}
+    {
+      cmd = "date -d \"" $3 "\" +%s 2>/dev/null || date -j -f %Y-%m-%d \"" $3 "\" +%s 2>/dev/null";
+      cmd | getline sent_epoch; close(cmd);
+      pending_days = (today - sent_epoch) / 86400;
+      if (pending_days >= days) {
+        printf "STALE REFERRAL: %s -> %s (%s) sent %s (%d days pending, no scheduled date)\n", $1, $2, $5, $3, pending_days;
+      }
+    }
+  ' "{{referrals_csv_path}}"
+variables:
+  referrals_csv_path:
+    default: "./specialist-referrals.csv"
+    description: "Path to a CSV with columns: pet_name,specialist_name,referral_sent_date (YYYY-MM-DD),scheduled_date (blank if not yet scheduled),specialty."
+  stale_days:
+    default: 14
+    description: Days since a referral was sent, with no scheduled date, before it's flagged.
+compatible_agents: [generic, claude, codex]
+notes: Assumes a maintained CSV export of outbound specialist referrals from your practice management system; adjust columns to match your export.