Mechanic

Mechanic — Vehicle Maintenance Tracker

Track mileage and service intervals for any combination of vehicles — trucks, cars, motorcycles, RVs, dirt bikes, ATVs, boats, and more. Decodes VINs to auto-populate vehicle specs, researches manufacturer-recommended maintenance schedules, tracks service history, estimates costs, monitors recalls, tracks fuel economy, manages warranties, and proactively reminds about upcoming and overdue services.

Data Storage

All user data lives in <workspace>/data/mechanic/:

Convention: Skill logic lives in <skill>/, user data lives in <workspace>/data/mechanic/. This keeps data safe when the skill is updated or reinstalled.

First-Time Setup

If <workspace>/data/mechanic/state.json doesn't exist:

1. Create <workspace>/data/mechanic/ directory
2. Ask the user what vehicles they want to track
3. For each vehicle, run the Adding a New Vehicle workflow (includes choosing check-in frequency per vehicle)
4. Create state.json with the vehicle entries
5. Set up the cron job (see Mileage Check Setup)

State File Structure

{
  "settings": {
    "check_in_tz": "America/Phoenix"
  },
  "providers": [
    {
      "id": "jims_diesel",
      "name": "Jim's Diesel Repair",
      "location": "123 Main St, Mesa, AZ",
      "phone": "480-555-1234",
      "specialties": ["diesel", "trucks"],
      "rating": 5,
      "notes": "Great with Power Stroke engines"
    }
  ],
  "vehicles": {
    "f350": {
      "label": "2021 Ford F-350 6.7L Power Stroke",
      "schedule_file": "f350-schedule.json",
      "check_in_frequency": "monthly",
      "current_miles": 61450,
      "last_updated": "2026-01-26",
      "last_check_in": "2026-01-26",
      "vin": "1FT8W3BT0MED12345",
      "vin_data": {
        "decoded": true,
        "decoded_date": "2026-01-26",
        "year": 2021,
        "make": "Ford",
        "model": "F-350",
        "trim": "Lariat",
        "body_class": "Pickup",
        "drive_type": "4WD",
        "engine": "6.7L Power Stroke V8 Turbo Diesel",
        "displacement_l": 6.7,
        "cylinders": 8,
        "fuel_type": "Diesel",
        "transmission": "10-Speed Automatic",
        "doors": 4,
        "gvwr_class": "Class 3",
        "bed_length": "8 ft",
        "wheel_base": "176 in",
        "plant_country": "United States",
        "plant_city": "Louisville",
        "raw_response": {}
      },
      "business_use": false,
      "business_use_percent": 0,
      "mileage_history": [
        {"date": "2026-01-26", "miles": 61450, "source": "user_reported"}
      ],
      "service_history": [
        {
          "service_id": "oil_filter",
          "date": "2025-11-15",
          "miles": 58000,
          "hours": null,
          "notes": "Full synthetic Motorcraft FL-2051S",
          "actual_cost": 125.00,
          "provider": {
            "id": "jims_diesel",
            "name": "Jim's Diesel Repair",
            "parts_warranty_months": 12,
            "labor_warranty_months": 6
          }
        }
      ],
      "fuel_history": [
        {
          "date": "2026-01-20",
          "gallons": 32.5,
          "cost": 108.55,
          "odometer": 61300,
          "mpg": 14.2,
          "notes": "Regular fill-up"
        }
      ],
      "warranties": [
        {
          "type": "factory_powertrain",
          "provider": "Ford",
          "start_date": "2021-03-15",
          "end_date": "2026-03-15",
          "start_miles": 0,
          "end_miles": 60000,
          "coverage_details": "Engine, transmission, drivetrain components",
          "status": "active"
        }
      ],
      "recalls": {
        "last_checked": "2026-01-26",
        "open_recalls": [],
        "completed_recalls": []
      },
      "emergency_info": {
        "vin": "1FT8W3BT0MED12345",
        "insurance_provider": "State Farm",
        "policy_number": "SF-123456789",
        "roadside_assistance_phone": "1-800-555-1234",
        "tire_size_front": "275/70R18",
        "tire_size_rear": "275/70R18",
        "tire_pressure_front_psi": 65,
        "tire_pressure_rear_psi": 80,
        "oil_type": "15W-40 CK-4 Full Synthetic",
        "oil_capacity": "15 quarts",
        "coolant_type": "Motorcraft Orange VC-3DIL-B",
        "def_type": "API certified DEF",
        "tow_rating_lbs": 20000,
        "gvwr_lbs": 14000,
        "gcwr_lbs": 37000,
        "key_fob_battery": "CR2450",
        "fuel_type": "Diesel (Ultra Low Sulfur)",
        "fuel_tank_gallons": 48,
        "notes": ""
      }
    }
  },
  "last_service_review": "2026-01-26"
}

Top-level fields:

• settings — global settings (timezone, etc.)
• providers — reusable list of service providers
• vehicles — keyed by short slug (e.g., f350, rv, crf450)
• last_service_review — date of last full review

Per-vehicle fields:

• label — human-readable vehicle name
• schedule_file — path to the service schedule JSON
• check_in_frequency — how often to ask for mileage (weekly/biweekly/monthly/quarterly)
• current_miles / current_hours — latest known readings
• last_updated / last_check_in — date tracking
• vin — Vehicle Identification Number (for recalls, VIN decode, and emergency info)
• vin_data — decoded VIN data from NHTSA VPIC API (specs, engine, transmission, etc.)
• business_use — whether vehicle is used for business (boolean)
• business_use_percent — percentage of business use (0-100)
• mileage_history — chronological array of mileage/hours entries
• service_history — chronological array of completed services (with optional actual_cost and provider)
• fuel_history — chronological array of fuel fill-ups
• warranties — array of warranty records
• recalls — recall monitoring state (last checked, open/completed)
• emergency_info — quick-reference vehicle specs and emergency contacts

Reading State

On skill load, read:

1. <workspace>/data/mechanic/state.json — current state for all vehicles
2. The relevant <key>-schedule.json file(s) depending on what's being discussed

Adding a New Vehicle

When the user wants to track a new vehicle:

1. Gather Vehicle Info

Ask for the VIN first. If the user provides a VIN, run the VIN Decode (see below) to auto-populate year, make, model, engine, transmission, drive type, and other specs. This saves the user from answering questions you can look up automatically.

Ask for:

• VIN (strongly recommended — auto-populates specs, enables recall monitoring, emergency info)
• Year, make, model (only ask if no VIN provided)
• Engine/trim (only ask if no VIN or VIN decode was incomplete)
• Usage pattern — daily driver, towing, off-road, weekend toy, etc.
• Current mileage/hours
• Business use? — if yes, what percentage? (enables tax deduction tracking)
• Warranty info — any active factory or extended warranties? Expiration date/mileage?
• Emergency info — insurance provider, roadside assistance number, tire sizes (can be filled in later)

If the user doesn't have the VIN handy, proceed with manual info and note that VIN can be added later to unlock auto-population and recall monitoring.

2. Determine Duty Level

Ask about usage to classify the maintenance schedule:

Most manufacturers publish both "normal" and "severe/special conditions" schedules. Use the one that matches.

3. Choose Check-In Frequency

Ask how often they want to be asked about this vehicle's mileage/hours:

Suggest a frequency based on the vehicle type and usage pattern, but let the user override.

4. Research the Service Schedule

Look up manufacturer-recommended maintenance intervals for that specific year/make/model/engine:

• Use web search to find the official maintenance schedule
• Check the owner's manual intervals
• Cross-reference with enthusiast forums for real-world advice
• Factor in the duty level from step 2

5. Build the Schedule File

Create <workspace>/data/mechanic/<key>-schedule.json:

{
  "vehicle": {
    "year": 2021,
    "make": "Ford",
    "model": "F-350",
    "type": "truck",
    "engine": "6.7L Power Stroke V8 Turbo Diesel",
    "transmission": "10R140 10-Speed Automatic",
    "duty": "severe",
    "notes": "Tows fifth wheel RV"
  },
  "services": [
    {
      "id": "oil_filter",
      "name": "Engine Oil & Filter Change",
      "interval_miles": 7500,
      "interval_months": 6,
      "details": "Specific oil type, filter part number, capacity, and any special instructions.",
      "priority": "critical",
      "cost_diy": "$XX-XX",
      "cost_shop": "$XX-XX",
      "cost_dealer": "$XX-XX",
      "cost_note": "Optional note about related expensive repairs"
    }
  ]
}

Required for every service item:

• id — unique snake_case identifier
• name — human-readable name
• At least one interval: interval_miles, interval_months, interval_hours, or interval_rides
• details — specific parts, fluids, capacities, and any warnings
• priority — critical, high, medium, or low
• cost_diy, cost_shop, cost_dealer — estimated cost ranges

Research costs:

• Search for typical costs for each service on that specific vehicle
• DIY = parts cost only
• Shop = independent mechanic
• Dealer = manufacturer dealership
• Add cost_note for items where failure/repair is significantly more expensive than maintenance

6. Add to State

Add the vehicle to state.json under the vehicles object:

{
  "<key>": {
    "label": "2021 Ford F-350 6.7L Power Stroke",
    "schedule_file": "<key>-schedule.json",
    "check_in_frequency": "monthly",
    "current_miles": 61450,
    "current_hours": null,
    "last_updated": "2026-01-26",
    "last_check_in": "2026-01-26",
    "vin": null,
    "vin_data": {
      "decoded": false
    },
    "business_use": false,
    "business_use_percent": 0,
    "mileage_history": [
      {"date": "2026-01-26", "miles": 61450, "source": "user_reported"}
    ],
    "service_history": [],
    "fuel_history": [],
    "warranties": [],
    "recalls": {
      "last_checked": null,
      "open_recalls": [],
      "completed_recalls": []
    },
    "emergency_info": {
      "vin": null,
      "insurance_provider": null,
      "policy_number": null,
      "roadside_assistance_phone": null,
      "tire_size_front": null,
      "tire_size_rear": null,
      "tire_pressure_front_psi": null,
      "tire_pressure_rear_psi": null,
      "oil_type": null,
      "oil_capacity": null,
      "coolant_type": null,
      "tow_rating_lbs": null,
      "gvwr_lbs": null,
      "key_fob_battery": null,
      "fuel_type": null,
      "fuel_tank_gallons": null,
      "notes": ""
    }
  }
}

Key naming: Use a short, memorable slug — f350, civic, r1, rv, crf450, harley, bass_boat, etc.

7. Update Cron Job

Update the cron job prompt to include the new vehicle. If this vehicle's frequency is higher than the current cron schedule, update the cron to fire at the higher frequency.

8. VIN Decode & Auto-Populate

If a VIN was provided, run the VIN Decode to auto-populate vehicle specs, emergency info fields, and the schedule file's vehicle section. Present the decoded info to the user for confirmation.

9. Run Initial Recall Check

If a VIN was provided, immediately check for open recalls (see NHTSA Recall Monitoring). If no VIN, check by make/model/year.

Vehicle Types and Special Considerations

Interval Types

Services can use any combination of:

• interval_miles — odometer-based
• interval_hours — engine/usage hours (generators, dirt bikes, boats)
• interval_months — time-based (everything degrades with age)
• interval_rides — per-use (e.g., dirt bike air filter = every ride)

Whichever interval is reached first triggers the service.

VIN Decode & Auto-Population

When a user provides a VIN (during vehicle setup or later), decode it using the free NHTSA VPIC API to automatically look up and store vehicle specifications.

NHTSA VPIC API (VIN Decoder)

Endpoint: https://vpic.nhtsa.dot.gov/api/vehicles/DecodeVinValues/{VIN}?format=json

No API key required. Free and unlimited.

Example:

GET https://vpic.nhtsa.dot.gov/api/vehicles/DecodeVinValues/1FT8W3BT0MED12345?format=json

Key Fields to Extract

The API returns a Results array with one object containing ~140+ fields. Extract and map these:

Note: Many fields return empty strings "" if not applicable. Only store non-empty values.

VIN Data Storage

Store decoded data in the vehicle's vin_data object in state.json:

{
  "vin_data": {
    "decoded": true,
    "decoded_date": "2026-01-27",
    "year": 2021,
    "make": "Ford",
    "model": "F-350",
    "trim": "Lariat",
    "body_class": "Pickup",
    "drive_type": "4WD",
    "engine": "6.7L Power Stroke V8 Turbo Diesel",
    "displacement_l": 6.7,
    "cylinders": 8,
    "fuel_type": "Diesel",
    "transmission": "10-Speed Automatic",
    "doors": 4,
    "gvwr_class": "Class 3",
    "bed_length": "8 ft",
    "wheel_base": "176 in",
    "plant_country": "United States",
    "plant_city": "Louisville",
    "raw_response": {}
  }
}

Store raw_response as the full VPIC result object for reference — it contains additional fields that may be useful later (e.g., AirBagLocFront, SeatBeltsAll, TPMS, ActiveSafetySysNote, etc.).

If vin_data.decoded is false or missing, the VIN hasn't been decoded yet.

Auto-Population Workflow

When a VIN is decoded:

1. Update vin_data — store all decoded fields
2. Update label — build from decoded year/make/model/engine (e.g., "2021 Ford F-350 6.7L Power Stroke")
3. Update emergency_info — auto-fill fields that can be derived:
   • fuel_type from FuelTypePrimary
   • gvwr_lbs from GVWR (parse weight class to approximate lbs)
4. Update schedule file — populate the vehicle section with decoded specs
5. Present to user — show what was decoded, confirm accuracy, ask about anything the VIN couldn't tell us (usage pattern, duty level, insurance, etc.)

When to Decode

Adding VIN Later

If a vehicle was added without a VIN and the user provides one later:

1. Decode the VIN
2. Store in vin_data
3. Update vin field
4. Backfill any empty emergency_info fields
5. Update label if the decoded info is more specific
6. Run an immediate recall check with the new VIN
7. Confirm what was updated

VIN Decode Presentation Format

When presenting decoded VIN data to the user:

🔍 VIN Decoded — [VIN]
━━━━━━━━━━━━━━━━━━━━━━━━━━━
📋 Vehicle
Year: [year] | Make: [make] | Model: [model]
Trim: [trim] | Body: [body_class]
Drive: [drive_type] | Doors: [doors]

🔧 Powertrain
Engine: [engine] ([displacement]L, [cylinders] cyl)
Fuel: [fuel_type]
Transmission: [transmission]

📏 Specs
GVWR: [gvwr_class]
Wheel Base: [wheel_base]
Bed Length: [bed_length] (if truck)

🏭 Built in [plant_city], [plant_country]
━━━━━━━━━━━━━━━━━━━━━━━━━━━

Limitations

• VPIC is NHTSA data — best for US-market vehicles. Import/foreign-market VINs may have incomplete data.
• Trailers and RVs — VIN decode may return limited data for trailers, fifth wheels, and RVs since they're built by different manufacturers with varying VIN encoding.
• Motorcycles and powersports — coverage varies. Japanese brands (Honda, Yamaha, Kawasaki, Suzuki) generally decode well. Smaller manufacturers may not.
• Pre-1981 vehicles — VINs weren't standardized until 1981. Older VINs won't decode.
• If decode returns sparse data, fall back to manual entry and web search for specs.

NHTSA Recall Monitoring

Monitor for open recalls on all tracked vehicles using the free NHTSA API (no API key required).

API Endpoints

• By make/model/year: https://api.nhtsa.dot.gov/recalls/recallsByVehicle?make=Ford&model=F-350&modelYear=2021
• By VIN (more precise): https://api.nhtsa.dot.gov/recalls/recallsByVin?vin=XXXXX

If a VIN is stored, prefer the VIN-based lookup. Otherwise fall back to make/model/year.

Recall Data Storage

Per vehicle in state.json:

{
  "recalls": {
    "last_checked": "2026-01-26",
    "open_recalls": [
      {
        "nhtsa_id": "26V-123",
        "component": "FUEL SYSTEM",
        "summary": "Fuel line may crack under pressure",
        "consequence": "Fuel leak, fire risk",
        "remedy": "Dealer will replace fuel line at no cost",
        "date_reported": "2025-12-01",
        "status": "open"
      }
    ],
    "completed_recalls": [
      {
        "nhtsa_id": "24V-456",
        "component": "ELECTRICAL",
        "summary": "Battery cable may corrode",
        "date_completed": "2025-06-15",
        "notes": "Done at dealer"
      }
    ]
  }
}

When to Check

• Monthly cron: Include recall checks in the mileage check cron. Check recalls for all vehicles monthly regardless of their check-in frequency.
• On vehicle add: Check immediately when a new vehicle is added.
• On demand: User asks "any recalls on my truck?"

Recall Report Format

Include in service review output:

⚠️ OPEN RECALLS
- [NHTSA ID] — [Component]: [Summary]
  Remedy: [What the dealer will do]
  ⚡ Contact your dealer to schedule this recall service (free)

When a user reports completing a recall, move it from open_recalls to completed_recalls with the completion date.

Fuel / MPG Tracking

Track fuel fill-ups to monitor fuel economy, detect mechanical issues early, and track fuel spending.

Logging a Fill-Up

When user says "filled up", "got gas/diesel", or reports a fill-up:

1. Capture: date, gallons, cost (total or price per gallon), odometer reading
2. Calculate MPG: (current_odometer - previous_odometer) / gallons
3. Append to the vehicle's fuel_history array
4. Check for MPG anomalies

Fuel History Entry

{
  "date": "2026-01-20",
  "gallons": 32.5,
  "cost": 108.55,
  "price_per_gallon": 3.34,
  "odometer": 61300,
  "mpg": 14.2,
  "partial_fill": false,
  "notes": ""
}

MPG Calculations

• Per fill-up MPG: (current_odometer - previous_fill_odometer) / gallons (skip if previous fill was partial)
• Rolling average: Average of last 10 fill-ups (or all if fewer than 10)
• Trend: Compare last 3 fill-ups to rolling average

Anomaly Detection

If a fill-up MPG is more than 15% below the rolling average, flag it:

⚠️ MPG Alert — [Vehicle]
Last fill-up: 10.5 MPG (your average is 14.2 MPG)
26% below your rolling average — this could indicate:
- Tire pressure issues
- Air filter needs replacement
- Fuel system issue
- Change in driving conditions (heavy towing, headwinds)
- Mechanical problem developing

Check tire pressures first, then review recent driving conditions.

Fuel Report Format

When asked "how's my fuel economy?" or "MPG report":

⛽ Fuel Report — [Vehicle]
Last fill-up: [X] MPG on [date]
Rolling average: [X] MPG (last 10 fills)
Trend: [improving/stable/declining]
Total fuel cost (YTD): $[X]
Total gallons (YTD): [X]
Average cost per gallon: $[X]

Partial Fills

If the user didn't fill up completely, mark partial_fill: true. Skip that entry for MPG calculation (the math won't be accurate), but still track cost and gallons.

Actual Cost Tracking

Track what the user actually pays for maintenance to build accurate spending records.

Capturing Costs

When a user logs a completed service:

1. After confirming the service details, ask: "What did you end up paying?" (or accept if they volunteered it)
2. Store as actual_cost in the service_history entry
3. If they don't know or don't want to share, leave it null — don't block the log

Service History Entry (with cost)

{
  "service_id": "oil_filter",
  "date": "2025-11-15",
  "miles": 58000,
  "hours": null,
  "notes": "Full synthetic, Motorcraft filter",
  "actual_cost": 125.00,
  "cost_type": "shop",
  "provider": {
    "id": "jims_diesel",
    "name": "Jim's Diesel Repair"
  }
}

cost_type values: diy, shop, dealer, warranty, recall (free)

Spending Analysis

Track and report on request:

• Per vehicle, per year: "You've spent $X on the F-350 this year"
• Actual vs estimated: Compare actual_cost to the schedule's cost estimates
• Category breakdown: Group by service type (oil changes, filters, tires, etc.)
• All-time total: Total maintenance spending per vehicle

Annual Summary

When asked or at year-end:

💰 [Year] Maintenance Summary — [Vehicle]
Total spent: $[X]
Services performed: [count]
Biggest expense: [service] — $[X]
Average cost per service: $[X]
vs. Estimated: $[X] ([over/under] by [X]%)

Warranty Tracking

Track warranties to know what's covered and get alerts before they expire.

Warranty Entry Structure

{
  "type": "factory_powertrain",
  "provider": "Ford",
  "start_date": "2021-03-15",
  "end_date": "2026-03-15",
  "start_miles": 0,
  "end_miles": 60000,
  "coverage_details": "Engine, transmission, transfer case, driveshaft, axle assemblies",
  "status": "active",
  "contact_phone": "1-800-392-3673",
  "claim_number": null,
  "notes": ""
}

Warranty Types

Expiration Alerts

Check warranties during every service review. Alert when:

• Within 3 months of end_date, OR
• Within 3,000 miles of end_miles (whichever comes first)

Alert format:

⚠️ WARRANTY EXPIRING SOON
[Vehicle] — [Warranty type] from [Provider]
Expires: [date] or [miles] miles (whichever first)
Remaining: ~[X] months / ~[X] miles
Coverage: [details]
💡 Schedule any warranty-covered concerns before expiration!

Warranty Coverage Check

When user asks "is this covered under warranty?" or when flagging a due service:

1. Check all active warranties for the vehicle
2. Match the service type to warranty coverage
3. If potentially covered: "This may be covered under your [warranty type] from [provider] (expires [date]). Contact them before paying out of pocket."

Status Values

• active — currently in effect
• expiring_soon — within alert threshold
• expired — past end date or end miles
• claimed — warranty claim was filed

Pre-Trip / Seasonal Checklists

Generate vehicle-specific checklists when a trip or seasonal change is mentioned.

Trigger Phrases

Activate when user says things like:

• "I'm heading on a trip" / "road trip" / "going to [location]"
• "towing this weekend" / "pulling the RV to [place]"
• "getting ready for winter" / "time to winterize"
• "spring is coming" / "time to de-winterize"
• "going off-road this weekend" / "trail ride"

Checklist Generation

Build the checklist by combining:

1. Overdue/due-soon services — Pull from service review for this vehicle
2. Weather at destination — Check forecast if location given (hot, cold, rain, snow, altitude)
3. Trip-type items — Based on what they're doing
4. Seasonal items — Based on current date and location

Towing Checklist (Truck + Trailer/RV)

🚛 Pre-Tow Checklist — [Truck] + [Trailer/RV]

TRUCK:
□ Engine oil level
□ Coolant level
□ DEF level (diesel)
□ Tire pressures (loaded spec: front [X] psi, rear [X] psi)
□ Brake controller connected and tested
□ Transmission temp gauge working
□ All lights working
□ Mirrors adjusted for towing

HITCH/CONNECTION:
□ Fifth wheel / gooseneck / ball mount secured
□ Pin box / kingpin locked (verify with tug test)
□ Safety chains crossed under tongue
□ Breakaway cable attached
□ 7-pin connector — test all lights (brake, turn, running, reverse)
□ Breakaway battery charged

TRAILER/RV:
□ Tire pressures (spec: [X] psi) — check age on sidewall
□ Wheel lug torque (spec: [X] ft-lbs)
□ Slides fully retracted and locked
□ Awning secured
□ Fridge set to travel mode (or propane off)
□ All compartments latched
□ Stabilizer jacks fully up
□ Roof vents closed
□ TV antenna down
□ Water heater bypass (if applicable)
□ LP gas tank valve position (check local laws for travel)
□ Cargo secured inside (open fridge, cabinets after arrival)

OVERDUE/DUE SERVICES:
[List any from service review]

Seasonal Checklists

Pre-Winter / Winterization:

• Antifreeze protection level (test with hydrometer)
• Battery load test (cold reduces capacity 30-50%)
• Wiper blades and washer fluid (cold-rated)
• Tire condition (all-season or winter tires?)
• Block heater working (diesel trucks)
• RV: Full winterization procedure (blow out lines, RV antifreeze, water heater drain, bypass)
• Boat: Winterize engine, fog cylinders, stabilize fuel, drain water systems

Pre-Summer / De-Winterization:

• AC system check (run it before you need it)
• Coolant level and condition
• RV: De-winterize water system, sanitize tanks, check AC units
• Check tire pressures (heat increases pressure)
• Inspect belts and hoses (heat accelerates wear)

Pre-Trip (General):

• All fluid levels
• Tire pressures and condition
• Lights and signals
• Brakes (visual or recent service)
• Wiper blades
• Emergency kit (jumper cables, flashlight, first aid)
• Registration and insurance current

Mileage Projection

Calculate driving pace and project when future services will come due.

Calculation

Requires 2+ data points in mileage_history at least 14 days apart.

average_miles_per_month = (latest_miles - earliest_miles) / months_between_readings

Use the full history for a stable average, but weight recent data more heavily if there's a significant change in driving pattern.

Service Date Projection

For each service:

1. Calculate miles remaining: next_due_miles - current_miles
2. Project months: miles_remaining / average_miles_per_month
3. Project date: today + projected_months
4. Also check time-based interval independently

Include in service review:

📅 Projected Service Dates
- Oil Change: ~[Month Year] (at ~[X] miles)
- Fuel Filters: ~[Month Year] (at ~[X] miles)
- Trans Fluid: ~[Month Year] (at ~[X] miles)

Budget Projection

When asked or included in reviews:

💰 Next 6-Month Budget Forecast — [Vehicle]
At [X] miles/month, expect:
- Oil change (~[Month]): $[X]
- Fuel filters (~[Month]): $[X]
- Cabin air filter (~[Month]): $[X]
Total estimated: $[X]

Insufficient Data

If fewer than 2 data points or readings are too close together:

• Note: "Need more mileage history to project dates — will be available after next check-in"
• Still show mile-based estimates without dates

Service Provider Tracking

Track where services are performed for easy reference and provider management.

Capturing Provider Info

When logging a completed service, optionally ask:

• Shop name (or "same as last time" / "DIY")
• Location (city/address)
• Phone number
• Satisfaction rating (1-5)
• Parts warranty (months)
• Labor warranty (months)

Don't make this burdensome — if they just say "got my oil changed", log the service first, then casually ask where. Skip if they seem uninterested.

Provider Storage

Providers are stored in two places:

1. Global providers array in state.json root — reusable across vehicles:

{
  "id": "jims_diesel",
  "name": "Jim's Diesel Repair",
  "location": "123 Main St, Mesa, AZ",
  "phone": "480-555-1234",
  "specialties": ["diesel", "trucks"],
  "rating": 5,
  "notes": "Great with Power Stroke engines"
}

2. Per service_history entry — reference by id plus any service-specific warranty:

{
  "provider": {
    "id": "jims_diesel",
    "name": "Jim's Diesel Repair",
    "parts_warranty_months": 12,
    "labor_warranty_months": 6
  }
}

Provider Queries

Handle questions like:

• "Where did I get my last oil change?" → Search service_history for most recent oil_filter entry, return provider
• "What shop did I use for the transmission service?" → Search by service_id
• "Show me all services at Jim's" → Filter service_history by provider.id
• "What's Jim's phone number?" → Look up in providers array
• "Same shop as last time" → Use the provider from the most recent service_history entry

Tax Deduction Integration

For vehicles flagged as business-use, help track deductible maintenance expenses.

Configuration

Per vehicle in state.json:

{
  "business_use": true,
  "business_use_percent": 50
}

If business_use is true and no percentage is set, assume 100%.

Deduction Tracking

When a service is completed with an actual_cost on a business-use vehicle:

1. Calculate deductible portion: actual_cost × (business_use_percent / 100)
2. Note it to the user:

   💼 Tax Note: This $450 trans fluid service is 50% business use.
   Deductible amount: $225.00 (vehicle maintenance expense)
   

3. Suggest logging to the tax-professional skill:

   Want me to log this to your tax deductions? 
   → $225.00 as vehicle maintenance expense
   

Integration with Tax-Professional Skill

If the user confirms, reference skills/tax-professional/SKILL.md and log to data/tax-professional/YYYY-expenses.json:

{
  "date": "2026-01-15",
  "category": "vehicle_maintenance",
  "description": "Trans fluid service — F-350 (50% business use)",
  "amount": 225.00,
  "vehicle": "f350",
  "full_cost": 450.00,
  "business_percent": 50,
  "receipt": false
}

Annual Tax Summary

On request or at tax time:

💼 [Year] Business Vehicle Deductions — [Vehicle]
Total maintenance costs: $[X]
Business use: [X]%
Deductible amount: $[X]
Services included: [count] services

Emergency Info Card

Store and quickly retrieve critical vehicle information for roadside emergencies, parts lookup, or quick reference.

Emergency Info Structure

Per vehicle in state.json:

{
  "emergency_info": {
    "vin": "1FT8W3BT0MED12345",
    "insurance_provider": "State Farm",
    "policy_number": "SF-123456789",
    "roadside_assistance_phone": "1-800-555-1234",
    "tire_size_front": "275/70R18",
    "tire_size_rear": "275/70R18",
    "tire_pressure_front_psi": 65,
    "tire_pressure_rear_psi": 80,
    "oil_type": "15W-40 CK-4 Full Synthetic",
    "oil_capacity": "15 quarts",
    "coolant_type": "Motorcraft Orange VC-3DIL-B",
    "def_type": "API certified DEF",
    "trans_fluid": "Motorcraft Mercon ULV",
    "tow_rating_lbs": 20000,
    "gvwr_lbs": 14000,
    "gcwr_lbs": 37000,
    "payload_lbs": 4300,
    "key_fob_battery": "CR2450",
    "fuel_type": "Diesel (Ultra Low Sulfur)",
    "fuel_tank_gallons": 48,
    "lug_nut_torque_ft_lbs": 165,
    "jack_points": "Frame rails, front and rear",
    "notes": ""
  }
}

Quick Access Queries

Respond instantly to:

• "What's my VIN?" → Return VIN
• "What are my truck's tire specs?" → Tire sizes and pressures
• "What oil does my truck take?" → Oil type and capacity
• "Insurance info?" → Provider, policy number, phone
• "Roadside assistance number?" → Phone number
• "What's my tow rating?" → Tow rating, GVWR, GCWR
• "Key fob battery?" → Battery type
• "Lug nut torque?" → Torque spec

Emergency Card Format

When asked for "emergency info" or "vehicle card":

🚨 Emergency Info — [Vehicle Label]
━━━━━━━━━━━━━━━━━━━━━━━━━━━
VIN: [vin]
Insurance: [provider] — Policy #[number]
Roadside: [phone]

🔧 Specs
Tires: F:[size] R:[size]
Pressure: F:[X]psi R:[X]psi
Oil: [type] ([capacity])
Coolant: [type]
Fuel: [type] ([tank] gal)
Key fob battery: [type]

📏 Ratings
Tow: [X] lbs | GVWR: [X] lbs
GCWR: [X] lbs | Payload: [X] lbs
Lug torque: [X] ft-lbs
━━━━━━━━━━━━━━━━━━━━━━━━━━━

Population

When adding a new vehicle, collect what's available. Many specs can be researched from the year/make/model. Let the user fill in personal info (insurance, roadside number) over time. Fields can be null until populated.

Cost Per Mile Analysis

Calculate total cost of vehicle ownership on a per-mile basis.

Calculations

Requires mileage history with at least 2 data points.

total_miles_driven = latest_miles - earliest_miles (from mileage_history)

Maintenance cost per mile = total_actual_costs / total_miles_driven
Fuel cost per mile = total_fuel_cost / total_miles_driven
Total operating cost per mile = (total_actual_costs + total_fuel_cost) / total_miles_driven

Report Format

When asked "cost per mile" or "operating cost":

📊 Cost Per Mile — [Vehicle]
Period: [earliest date] to [latest date] ([X] miles driven)

Maintenance only: $[X.XX]/mile
Fuel only: $[X.XX]/mile (if fuel tracking active)
Total operating: $[X.XX]/mile

💡 National averages (approximate):
Cars: ~$0.10/mi maintenance, ~$0.12/mi fuel
Trucks: ~$0.14/mi maintenance, ~$0.20/mi fuel (diesel)
Heavy-duty diesel (towing): ~$0.18/mi maintenance, ~$0.25/mi fuel

Fleet Overview

If tracking multiple vehicles, show a comparison:

📊 Fleet Cost Per Mile
[Vehicle 1]: $[X.XX]/mi (maintenance) | $[X.XX]/mi (total)
[Vehicle 2]: $[X.XX]/mi (maintenance) | $[X.XX]/mi (total)
Fleet average: $[X.XX]/mi

Notes

• Only include services with actual_cost recorded (skip nulls)
• If no fuel data, show maintenance-only
• Warn if data period is short (<3 months or <1,000 miles): "Limited data — will become more accurate over time"
• Cost per mile naturally decreases as expensive one-time services are amortized over more miles

Mileage Check (Cron-Triggered)

A single cron job runs weekly (the highest possible frequency) and checks which vehicles are due for a check-in based on their individual check_in_frequency and last_check_in date. It also performs monthly recall checks.

Prompt: "Mechanic skill: mileage check"

When this fires:

1. Read <workspace>/data/mechanic/state.json
2. For each vehicle, check if it's due for a check-in:
   • Compare last_check_in date against check_in_frequency
   • Weekly: due if 7+ days since last check-in
   • Biweekly: due if 14+ days
   • Monthly: due if 30+ days
   • Quarterly: due if 90+ days
3. Monthly recall check: If 30+ days since any vehicle's recalls.last_checked, fetch latest recalls from NHTSA API and update state
4. If no vehicles are due for check-in AND no new recalls found, reply with HEARTBEAT_OK (skip silently)
5. If vehicles are due, ask for current readings on only the due vehicles
6. If new recalls found, include them in the message even if no vehicles are due for mileage check-in
7. Update state and last_check_in when they respond
8. Run a Service Review for the updated vehicles (includes warranty alerts, projections, recalls)

Mileage Check Setup

Create a

…(truncated)