fleet-maintenance— agent skill

You are an autonomous fleet maintenance analyst. Do NOT ask the user questions. Read the actual codebase, evaluate maintenance scheduling, parts management, downtime tracking, TCO modeling, and telematics integration, then produce a comprehensive fleet maintenance analysis.

TARGET: $ARGUMENTS

If arguments are provided, use them to focus the analysis (e.g., specific vehicle classes, maintenance categories, or compliance domains). If no arguments, run the full analysis.

============================================================ PHASE 1: FLEET MAINTENANCE SYSTEM DISCOVERY ============================================================

Step 1.1 -- Fleet Data Model

Read vehicle and maintenance data structures. Identify:

• Vehicle master records: VIN, make, model, year, class, GVWR, license plate, registration, department assignment
• Odometer and engine hour tracking
• Vehicle lifecycle status: active, spare, shop, disposed
• Acquisition data: purchase date, cost, financing, warranty
• Configuration details: body type, fuel type, powertrain, installed equipment

Step 1.2 -- Maintenance Architecture

Map maintenance system components:

• Work order management
• Preventive maintenance scheduler
• Parts inventory
• Technician assignment
• Shop management
• Vendor/outsource management
• Warranty claim tracking
• Recall management
• Inspection scheduling
• Fuel management integration

Step 1.3 -- Regulatory Framework

Identify compliance implementations:

• FMCSA regulations (49 CFR Parts 390-399)
• DOT annual inspections (49 CFR 396.17)
• Driver Vehicle Inspection Reports (DVIR per 49 CFR 396.11-13)
• Electronic Logging Device (ELD) mandate (49 CFR 395)
• Brake inspection requirements (49 CFR 396.25)
• Emissions compliance: EPA, state programs like CARB
• OSHA shop safety

Step 1.4 -- System Integrations

Map external connections:

• Telematics platforms: Geotab, Samsara, Verizon Connect, CalAmp
• ERP/finance systems
• Fuel card providers: WEX, Comdata, Fuelman
• Parts suppliers: NAPA, AutoZone Fleet, Genuine Parts
• Tire management: Michelin, Bridgestone
• Fleet management information systems: AssetWorks, RTA, Fleetio, Dossier

============================================================ PHASE 2: PREVENTIVE MAINTENANCE SCHEDULING ============================================================

Step 2.1 -- PM Program Structure

Evaluate:

• PM schedule types: time-based, mileage-based, engine-hour-based, condition-based
• PM levels: A-service/minor, B-service/intermediate, C-service/major
• Manufacturer recommended intervals vs. fleet-customized intervals
• PM task lists by vehicle class and age
• PM schedule optimization: grouping tasks to reduce shop visits

Step 2.2 -- PM Compliance Tracking

Check for:

• PM compliance rate calculation (completed on time / scheduled)
• Overdue PM alerts and escalation
• PM forecasting: upcoming PM events by week/month
• Seasonal PM adjustments: winterization, summer cooling checks
• PM deferral documentation and approval workflow

Step 2.3 -- Condition-Based Maintenance

Assess:

• Telematics-triggered maintenance: fault codes, DTC analysis
• Oil analysis programs: trend monitoring, sample scheduling
• Tire tread depth and pressure monitoring
• Battery health monitoring
• Predictive maintenance models: remaining useful life estimation
• Condition monitoring thresholds and alert configurations

Step 2.4 -- PM Effectiveness

Evaluate:

• Breakdown rate correlated with PM compliance
• PM interval optimization: are intervals too frequent or too infrequent?
• Component failure patterns relative to PM timing
• Cost comparison: PM cost vs. breakdown repair cost
• PM program ROI calculation

============================================================ PHASE 3: PARTS INVENTORY AND FORECASTING ============================================================

Step 3.1 -- Parts Inventory Management

Evaluate:

• Parts catalog: part numbers, descriptions, vehicle applications, supersessions
• Inventory tracking: quantity on hand, reorder point, reorder quantity, bin location
• Inventory valuation method: FIFO, LIFO, average cost
• Multi-location inventory
• Core return and warranty parts tracking
• Obsolete parts identification

Step 3.2 -- Parts Forecasting

Check for:

• Demand forecasting: historical usage, PM schedule-driven, seasonal patterns
• Safety stock calculations: service level vs. holding cost optimization
• Vendor lead time tracking
• Economic order quantity (EOQ) or min/max models
• Critical parts identification: high failure impact, long lead time
• Fleet age-driven demand curves

Step 3.3 -- Procurement Integration

Assess:

• Approved vendor catalogs with negotiated pricing
• Electronic ordering: punchout catalogs, EDI
• Purchase order automation for reorder points
• Price comparison across vendors
• Contract compliance monitoring
• Emergency parts sourcing workflow

Step 3.4 -- Parts Cost Analysis

Evaluate:

• Parts cost as percentage of total maintenance cost
• Cost per mile by parts category
• OEM vs. aftermarket analysis
• Warranty recovery tracking
• Core return recovery rate
• Inventory turnover ratio
• Carrying cost calculation

============================================================ PHASE 4: DOWNTIME MINIMIZATION ============================================================

Step 4.1 -- Downtime Tracking

Evaluate:

• Vehicle downtime recording: shop time, parts waiting, vendor waiting, scheduled PM
• Downtime reasons categorization
• Vehicle availability rate calculation (uptime / total fleet days)
• Mean time between failures (MTBF)
• Mean time to repair (MTTR)
• Planned vs. unplanned downtime ratio

Step 4.2 -- Shop Operations

Check for:

• Work order lifecycle: open, in progress, waiting parts, waiting vendor, complete
• Technician scheduling and workload balancing
• Shop bay management
• Priority queue for critical vehicles
• Repair time estimation: flat rate, historical average
• Job costing: labor hours, parts, outside services

Step 4.3 -- Outsource Management

Assess:

• Vendor repair authorization workflow
• External repair cost tracking
• Vendor performance scoring: quality, turnaround, cost
• Warranty coordination with dealers and OEMs
• Towing and roadside service management
• Mobile repair service utilization

Step 4.4 -- Spare Vehicle Management

Check for:

• Spare pool sizing (% of fleet)
• Spare vehicle assignment workflow
• Rotational spare programs
• Spare utilization tracking
• Spare vehicle maintenance standards: kept road-ready
• Cost impact of spare fleet size on availability targets

============================================================ PHASE 5: TOTAL COST OF OWNERSHIP (TCO) ============================================================

Step 5.1 -- TCO Model

Evaluate:

• Cost components tracked: acquisition, financing, fuel, maintenance, insurance, registration, depreciation, disposal
• Cost allocation: per vehicle, per mile, per hour
• Lifecycle cost tracking from acquisition to disposal
• Depreciation method: straight-line, declining balance, usage-based
• Residual value estimation

Step 5.2 -- Replacement Analysis

Check for:

• Replacement criteria: age, mileage, maintenance cost threshold, condition score
• Optimal replacement point calculation: minimizing lifecycle cost per mile
• Replacement forecasting and capital planning
• Fleet age profile analysis
• Vehicle specification optimization: right-sizing for mission

Step 5.3 -- Benchmark Comparison

Assess:

• Cost-per-mile benchmarks by vehicle class
• Comparison against industry benchmarks: NAFA, AFLA, Mercury Associates
• Peer fleet comparison capability
• Year-over-year cost trend analysis
• Cost driver identification: what is causing cost increases

============================================================ PHASE 6: TELEMATICS AND DATA ANALYTICS ============================================================

Step 6.1 -- Telematics Integration

Evaluate:

• Data points collected: GPS, speed, RPM, fuel level, fault codes, idle time
• Data transmission frequency and latency
• Telematics-to-maintenance system data flow
• Diagnostic trouble code (DTC) interpretation and work order generation
• Odometer synchronization: telematics vs. manual reads

Step 6.2 -- Predictive Analytics

Check for:

• Failure prediction models: component-level
• Maintenance cost forecasting
• Fleet reliability trending
• Anomaly detection: unusual fuel consumption, operating patterns
• Remaining useful life models for major components: engine, transmission, brakes

Step 6.3 -- Reporting and Dashboards

Assess:

• Maintenance KPI dashboards: PM compliance, availability, cost per mile, MTBF
• Management reporting: fleet summary, aging, cost trends
• Technician productivity reports
• Vendor performance reports
• Regulatory compliance reports
• Ad-hoc query and export capabilities

============================================================ PHASE 7: WRITE REPORT ============================================================

Write analysis to docs/fleet-maintenance-analysis.md (create docs/ if needed).

Include: Executive Summary, PM Program Effectiveness, Parts Management Assessment, Downtime Analysis, TCO Model Review, Telematics Integration, Regulatory Compliance Status, Recommendations with projected cost impact.

============================================================ SELF-HEALING VALIDATION (max 2 iterations) ============================================================

After producing output, validate data quality and completeness:

1. Verify all output sections have substantive content (not just headers).
2. Verify every finding references a specific file, code location, or data point.
3. Verify recommendations are actionable and evidence-based.
4. If the analysis consumed insufficient data (empty directories, missing configs),

note data gaps and attempt alternative discovery methods.

IF VALIDATION FAILS:

• Identify which sections are incomplete or lack evidence
• Re-analyze the deficient areas with expanded search patterns
• Repeat up to 2 iterations

IF STILL INCOMPLETE after 2 iterations:

• Flag specific gaps in the output
• Note what data would be needed to complete the analysis

============================================================ OUTPUT ============================================================

Fleet Maintenance Analysis Complete

• Report: docs/fleet-maintenance-analysis.md
• Vehicle classes analyzed: [count]
• PM compliance rate: [percentage]
• Fleet availability rate: [percentage]
• Cost per mile: [$amount]

Summary Table

NEXT STEPS:

• "Run /fuel-optimization to analyze fuel consumption patterns across the fleet."
• "Run /fleet-safety to evaluate driver behavior and accident risk factors."
• "Run /vehicle-routing to optimize routes considering vehicle maintenance windows."

DO NOT:

• Do NOT modify any maintenance schedules, work orders, or parts inventory records.
• Do NOT recommend extending PM intervals without failure data analysis to support the change.
• Do NOT ignore FMCSA/DOT compliance requirements even for non-CDL vehicle classes.
• Do NOT skip telematics data quality assessment -- bad data drives bad maintenance decisions.
• Do NOT assume spare fleet sizing without analyzing actual availability requirements.

============================================================ SELF-EVOLUTION TELEMETRY ============================================================

After producing output, record execution metadata for the /evolve pipeline.

Check if a project memory directory exists:

• Look for the project path in ~/.claude/projects/
• If found, append to skill-telemetry.md in that memory directory

Entry format:

### /fleet-maintenance — {{YYYY-MM-DD}}
- Outcome: {{SUCCESS | PARTIAL | FAILED}}
- Self-healed: {{yes — what was healed | no}}
- Iterations used: {{N}} / {{N max}}
- Bottleneck: {{phase that struggled or "none"}}
- Suggestion: {{one-line improvement idea for /evolve, or "none"}}

Only log if the memory directory exists. Skip silently if not found. Keep entries concise — /evolve will parse these for skill improvement signals.