Prompts • Skills • Connectors • Agents • Logistics workflows • Tools
AI Playbook
for Logistics & Transportation Companies
For operations, dispatch, and finance leaders at 3PLs, freight brokers, and trucking fleets. A practical guide to load planning, carrier sourcing, freight billing and audit, detention and accessorials, DOT and hours-of-service compliance, claims, and customer quotes.
Who this playbook is for
This playbook is for 3PLs, brokers, and fleets, not owner-operators. It's written for teams with dispatchers, carrier sales reps, billing staff, and safety and compliance managers. It doesn't cover running a single-truck business, and it isn't for shippers managing their own freight.
- Third-party logistics providers (3PLs)
- Freight brokerages
- Trucking fleets
- Operations, dispatch, and carrier sales leaders
- Billing, finance, safety, compliance, and claims teams
- Owner-operators running their own truck
- Shippers managing their own freight
- Drivers looking for job or route advice
Start With Your Role
Choose the role closest to your day-to-day work. Start with one useful prompt, make it repeatable, then add the information it needs.
- Start with: daily load board review
- Save as a skill: risk rules and handoff format
- Connect: TMS exports, read-only telematics
- First agent: morning load risk board
- Measure: on-time %, uncovered loads
- Start with: lane carrier shortlist or fraud red-flag review
- Save as a skill: shortlist criteria and vetting checklist
- Connect: carrier history, vetting platform reports
- First agent: carrier fraud red-flag screener
- Measure: time to cover, fraud incidents
- Start with: spot quote builder or contract rate review
- Save as a skill: margin rules and quote format
- Connect: market rate exports, lane history
- First agent: spot quote assistant
- Measure: win rate, margin
- Start with: carrier invoice audit or billing readiness
- Save as a skill: audit rules and billing requirements
- Connect: TMS billing, document management
- First agent: carrier invoice auditor
- Measure: overpayments prevented, DSO
- Start with: HOS violation trend review or DQ file audit
- Save as a skill: compliance checklists
- Connect: Samsara MCP (read-only) or ELD reports
- First agent: HOS pattern watch
- Measure: violations, audit findings
- Start with: claim intake summary
- Save as a skill: claim file format and deadlines
- Connect: documents, temperature logs
- First agent: claims deadline tracker
- Measure: cycle time, recovery
100 AI Prompts for Logistics & Transportation
Ready-to-use prompts for load planning, carrier sourcing, vetting and fraud prevention, quotes, billing and audit, detention and accessorials, DOT and HOS compliance, safety and maintenance, claims, and finance. Copy one, add your operation's context, and review the result.
Use your TMS and telematics data, quote rate confirmations and contracts exactly, and review the result before it reaches a driver, carrier, or customer.
Prompts for planning loads, matching equipment and drivers, and managing the daily board. Dispatchers make the assignments.
One clear risk-sorted list starts the dispatch day focused.
Risks: Status data may lag. Control: Dispatchers confirm with drivers or carriers before acting.
Shows feasible options quickly, including hours-of-service fit.
Risks: HOS feasibility depends on live ELD data. Control: Dispatcher confirms hours with the driver and ELD before assigning.
Sequencing stops around windows reduces detention and missed appointments.
Risks: AI isn't a truck-routing engine. Control: Use your routing software for truck-legal routes and miles.
Deadhead is pure cost; backhauls turn it into revenue.
Risks: Rates on backhauls must still make sense. Control: Pricing approves.
Seeing gaps early lets you secure capacity at better rates.
Risks: Forecasts change. Control: Update daily.
Missed appointments cause detention, rework, and service failures.
Risks: Facility rules vary. Control: Coordinator books in each portal.
Proactive, consistent updates protect customer relationships.
Risks: ETAs must be confirmed. Control: Rep verifies ETA before sending.
Clean handoffs prevent dropped issues.
Risks: Notes must be complete. Control: Outgoing lead reviews.
Shows where planning improvements pay off.
Risks: Late-cause coding varies. Control: Standardize codes.
Lost or idle trailers are expensive capital.
Risks: Customer contracts govern pools. Control: Account manager coordinates.
A plan made before the storm saves loads and keeps drivers safe.
Risks: Weather changes. Control: Safety and dispatch leaders decide holds; drivers can always stop when unsafe.
12 Claude-Ready Logistics Skills
These are downloadable Claude Skill packages, not just longer prompts. Each includes a SKILL.md your team can tailor with your rules, customers, and contracts. A copy-and-paste version works in ChatGPT and other AI tools.
Connect Your Work
AI can only help with the information it can see. Start with one system, see what it can do, then add more when you are ready.
Easy start
Start with a limited, low-risk connection your team can test quickly.
- Access: Samsara's generally available MCP server with read-only tools for vehicles, drivers, HOS, DVIRs, routes, dispatch, fuel, and locations; works with Claude, ChatGPT Enterprise, and Microsoft Copilot Studio
- Useful for: HOS patterns, fleet location questions, and maintenance checks
- Setup & limit: read-only; access mirrors each user's Samsara permissions; the org owner adds it in Claude; use your region's endpoint
- Access: Gmail, Drive, Sheets, and Calendar connections in Claude and ChatGPT
- Useful for: rate confirmations, SOPs, and customer files
- Setup & limit: start with operations folders; keep driver personal data out
- Access: Outlook, SharePoint, Excel, and Teams connections, or Microsoft Copilot in your tenant
- Useful for: SOPs, contracts, and reports
- Setup & limit: admin controls access; start with one site
- Access: official Slack connection for Claude and ChatGPT; Teams via Microsoft 365
- Useful for: dispatch and operations channel summaries
- Setup & limit: exclude HR and driver disciplinary channels
Needs an administrator
Useful, but someone needs to set access and permissions first.
- Access: Motive's MCP connector for Claude and ChatGPT can query fleet, HOS, safety, and dispatch data and take actions (send driver messages, create dispatches, update vehicles)
- Useful for: fleet questions and, with approval, specific actions
- Setup & limit: only fleet admin accounts can connect today, and connected users see fleet-wide data; keep actions disabled in your process unless an owner approves
- Access: Geotab offers an AI connector for fleet data
- Useful for: maintenance and fleet utilization questions
- Setup & limit: admin sets up access; review data scope
- Access: QuickBooks, NetSuite (AI Connector Service with admin setup), or your ERP
- Useful for: AR, margin, and cost questions
- Setup & limit: read-only role; postings and payments stay with finance
Integration project
Plan the use case, source data, permissions, and owner before connecting.
- Access: no official direct Claude or ChatGPT connection is confirmed here for most TMS platforms; many offer APIs or reporting exports
- Useful for: scheduled exports of load board, billing, and carrier data
- Setup & limit: define fields and owner; avoid write access to loads and invoices
- Access: DAT and Truckstop provide rate and capacity data under their licenses
- Useful for: market rate exports for quoting and benchmarks
- Setup & limit: follow data license terms
- Access: Highway, Carrier Assure, RMIS, and MyCarrierPortal
- Useful for: vetting reports and monitoring alerts
- Setup & limit: vetting decisions stay in the platform with compliance staff
Good to know
Limits and honest gaps.
- Access: Clearinghouse results, test results, and medical certificates
- Useful for: nothing in AI tools; track only dates and completion status
- Setup & limit: the DER manages these records
- Access: single-truck owner-operators have different needs
- Useful for: not covered here; this playbook is for 3PLs, brokers, and fleets
- Setup & limit: fleets that lease on owner-operators can use the fleet sections
Tools
Choose the AI tools and logistics software that fit your operation. The goal is to make your existing systems easier to use, not replace everything at once.
AI Assistants
4TMS
7Capacity & Market Data
5Carrier Vetting
4Billing, Audit & Payments
5Safety & Compliance
4Collaboration
312 AI Agents for Logistics & Transportation
An agent is an assistant for one recurring job. It gathers the right information, follows your process, and prepares the work for someone to review and decide.
A good first agent handles one task your team already does, reads data without changing it, never messages drivers, carriers, or customers, and hands its work to a named person.
# Morning Load Risk Board ## Job to be done Start every dispatch shift with a clear list of loads at risk. ## When it runs Start of each dispatch shift; when a load status changes to late. ## Approved context TMS load board, tracking status, appointments, driver availability with ELD hours. ## Operating loop Identify uncovered and at-risk loads; list feasible options with HOS fit; flag missing appointments; draft customer update drafts. ## What it delivers A risk-sorted dispatch list. ## Human owner Dispatch manager. ## What it must never do Never assign loads, message drivers, or plan beyond HOS limits. ## How to measure it On-time pickup and delivery, uncovered loads at cutoff. ## Operating rules 1. Use only approved sources and follow the linked logistics skill. 2. Cite load, invoice, or record IDs and separate facts from assumptions. 3. Route safety risks, fraud indicators, HOS conflicts, and deadline risks to the human owner immediately. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when regulation, contract terms, or approval authority are unclear. 6. Use telematics and TMS connectors read-only unless the owner has approved a specific action.
# Lane Capacity Finder ## Job to be done Find good carriers fast for uncovered and recurring lanes. ## When it runs When a load is uncovered within 24 hours of pickup; weekly for recurring lanes. ## Approved context Carrier history, scorecards, market rate exports. ## Operating loop Rank carriers by lane history and performance; check vetting status; draft outreach for the rep. ## What it delivers A shortlist per lane. ## Human owner Carrier sales manager. ## What it must never do Never tender loads or quote rates without approval. ## How to measure it Time to cover, cost vs. market, carrier performance. ## Operating rules 1. Use only approved sources and follow the linked logistics skill. 2. Cite load, invoice, or record IDs and separate facts from assumptions. 3. Route safety risks, fraud indicators, HOS conflicts, and deadline risks to the human owner immediately. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when regulation, contract terms, or approval authority are unclear. 6. Use telematics and TMS connectors read-only unless the owner has approved a specific action.
# Carrier Fraud Red-Flag Screener ## Job to be done Catch fraud indicators before a load is tendered. ## When it runs When a new carrier is set up; when carrier contact or payment info changes; when monitoring alerts arrive. ## Approved context Carrier setup data, FMCSA pulls, COI, monitoring alerts. ## Operating loop Compare details to requirements; list red flags; flag active loads with affected carriers; list verification steps. ## What it delivers A red-flag packet for compliance. ## Human owner Carrier compliance manager. ## What it must never do Never approve or reject carriers or contact them. ## How to measure it Fraud incidents, time to vet, carriers suspended appropriately. ## Operating rules 1. Use only approved sources and follow the linked logistics skill. 2. Cite load, invoice, or record IDs and separate facts from assumptions. 3. Route safety risks, fraud indicators, HOS conflicts, and deadline risks to the human owner immediately. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when regulation, contract terms, or approval authority are unclear. 6. Use telematics and TMS connectors read-only unless the owner has approved a specific action.
# Spot Quote Assistant ## Job to be done Speed up spot quotes with consistent, data-based ranges. ## When it runs When a quote request arrives. ## Approved context Market rate data, our lane history, fuel tables, margin targets. ## Operating loop Calculate buy and sell ranges; list accessorials to quote; draft quote email. ## What it delivers A quote recommendation and draft. ## Human owner Pricing rep. ## What it must never do Never send quotes. ## How to measure it Quote turnaround, win rate, margin. ## Operating rules 1. Use only approved sources and follow the linked logistics skill. 2. Cite load, invoice, or record IDs and separate facts from assumptions. 3. Route safety risks, fraud indicators, HOS conflicts, and deadline risks to the human owner immediately. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when regulation, contract terms, or approval authority are unclear. 6. Use telematics and TMS connectors read-only unless the owner has approved a specific action.
# Billing Readiness Monitor ## Job to be done Get delivered loads billed fast. ## When it runs Daily. ## Approved context Delivered loads, documents received, customer billing rules. ## Operating loop List loads ready to bill and missing documents; draft document requests to carriers for the team. ## What it delivers A billing readiness list. ## Human owner Billing manager. ## What it must never do Never send invoices or contact customers. ## How to measure it Days delivered-to-billed, unbilled revenue. ## Operating rules 1. Use only approved sources and follow the linked logistics skill. 2. Cite load, invoice, or record IDs and separate facts from assumptions. 3. Route safety risks, fraud indicators, HOS conflicts, and deadline risks to the human owner immediately. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when regulation, contract terms, or approval authority are unclear. 6. Use telematics and TMS connectors read-only unless the owner has approved a specific action.
# Carrier Invoice Auditor ## Job to be done Catch carrier invoice errors and duplicates before payment. ## When it runs When carrier invoices arrive. ## Approved context Invoices, rate confirmations, approvals, POD, factoring notices. ## Operating loop Compare invoices to rate confirmations; flag unapproved accessorials and duplicates; total variances. ## What it delivers An audit exception queue. ## Human owner AP manager. ## What it must never do Never approve, short-pay, or pay invoices, or change payment details. ## How to measure it Overpayments prevented, duplicates caught. ## Operating rules 1. Use only approved sources and follow the linked logistics skill. 2. Cite load, invoice, or record IDs and separate facts from assumptions. 3. Route safety risks, fraud indicators, HOS conflicts, and deadline risks to the human owner immediately. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when regulation, contract terms, or approval authority are unclear. 6. Use telematics and TMS connectors read-only unless the owner has approved a specific action.
# Detention Capture Assistant ## Job to be done Bill the detention you're owed with solid evidence. ## When it runs When a load delivers with dwell over free time. ## Approved context ELD or geofence timestamps, appointments, contract terms. ## Operating loop Calculate billable detention; assemble evidence; draft the customer note and carrier pay record. ## What it delivers A detention package. ## Human owner Billing specialist. ## What it must never do Never bill customers or pay carriers. ## How to measure it Detention billed and collected, dispute rate. ## Operating rules 1. Use only approved sources and follow the linked logistics skill. 2. Cite load, invoice, or record IDs and separate facts from assumptions. 3. Route safety risks, fraud indicators, HOS conflicts, and deadline risks to the human owner immediately. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when regulation, contract terms, or approval authority are unclear. 6. Use telematics and TMS connectors read-only unless the owner has approved a specific action.
# HOS Pattern Watch ## Job to be done Spot HOS violation patterns, including those caused by planning. ## When it runs Weekly; daily for unassigned driving. ## Approved context ELD reports (Samsara MCP is read-only; Motive MCP requires a fleet admin), dispatch records. ## Operating loop Summarize violations and unassigned driving; link patterns to terminals and dispatch; list follow-ups. ## What it delivers An HOS compliance report. ## Human owner Safety and compliance manager. ## What it must never do Never edit logs, assign driving time, or message drivers. ## How to measure it Violations per 1,000 hours, unassigned driving resolved. ## Operating rules 1. Use only approved sources and follow the linked logistics skill. 2. Cite load, invoice, or record IDs and separate facts from assumptions. 3. Route safety risks, fraud indicators, HOS conflicts, and deadline risks to the human owner immediately. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when regulation, contract terms, or approval authority are unclear. 6. Use telematics and TMS connectors read-only unless the owner has approved a specific action.
# DQ File & Compliance Calendar Monitor ## Job to be done Keep every driver file and filing current. ## When it runs Weekly; 60 and 30 days before expirations or filing deadlines. ## Approved context DQ file status, compliance calendar. ## Operating loop List expiring items and drivers not dispatchable until resolved; draft reminders; flag filing deadlines. ## What it delivers A compliance status report. ## Human owner Compliance manager. ## What it must never do Never mark items complete or access drug and alcohol results. ## How to measure it Expired items, audit findings, on-time filings. ## Operating rules 1. Use only approved sources and follow the linked logistics skill. 2. Cite load, invoice, or record IDs and separate facts from assumptions. 3. Route safety risks, fraud indicators, HOS conflicts, and deadline risks to the human owner immediately. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when regulation, contract terms, or approval authority are unclear. 6. Use telematics and TMS connectors read-only unless the owner has approved a specific action.
# PM & Defect Tracker ## Job to be done Make sure no unit runs overdue for PM or with open safety defects. ## When it runs Daily. ## Approved context PM schedule, DVIR defects, odometer data. ## Operating loop List units due, overdue, or with open defects; suggest shop schedule. ## What it delivers A daily maintenance list. ## Human owner Shop manager. ## What it must never do Never clear defects or release units to service. ## How to measure it PM compliance, roadside OOS rate, breakdowns. ## Operating rules 1. Use only approved sources and follow the linked logistics skill. 2. Cite load, invoice, or record IDs and separate facts from assumptions. 3. Route safety risks, fraud indicators, HOS conflicts, and deadline risks to the human owner immediately. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when regulation, contract terms, or approval authority are unclear. 6. Use telematics and TMS connectors read-only unless the owner has approved a specific action.
# Claims Deadline Tracker ## Job to be done Protect rights and resolve claims on time. ## When it runs Daily for open claims. ## Approved context Claim files, deadlines, documentation status. ## Operating loop List claims near deadlines; list missing documents; draft requests; update aging summary. ## What it delivers A claims status report. ## Human owner Claims manager. ## What it must never do Never accept, deny, or settle claims. ## How to measure it Claims past deadline, cycle time, recovery. ## Operating rules 1. Use only approved sources and follow the linked logistics skill. 2. Cite load, invoice, or record IDs and separate facts from assumptions. 3. Route safety risks, fraud indicators, HOS conflicts, and deadline risks to the human owner immediately. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when regulation, contract terms, or approval authority are unclear. 6. Use telematics and TMS connectors read-only unless the owner has approved a specific action.
# Customer Margin & AR Watch ## Job to be done Spot customers with falling margin or slowing payment. ## When it runs Weekly. ## Approved context Load revenue and cost, accessorials, AR aging, credit limits. ## Operating loop Calculate margin by customer; flag declines, accessorial leakage, slow pay, and over-limit accounts. ## What it delivers A weekly customer finance report. ## Human owner Finance manager and account leaders. ## What it must never do Never change credit limits or contact customers. ## How to measure it Margin trend, DSO, bad debt. ## Operating rules 1. Use only approved sources and follow the linked logistics skill. 2. Cite load, invoice, or record IDs and separate facts from assumptions. 3. Route safety risks, fraud indicators, HOS conflicts, and deadline risks to the human owner immediately. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when regulation, contract terms, or approval authority are unclear. 6. Use telematics and TMS connectors read-only unless the owner has approved a specific action.
AI for Load Planning & Dispatch
Use AI to sort the load board, show feasible options, and plan capacity. Dispatchers make assignments and drivers' safety comes first.
- Prompts: daily load board review, driver-load matching, weekly capacity plan
- Skills: load board & dispatch planning
- Connect: TMS exports, Samsara MCP (read-only) or Motive MCP (fleet admin), appointment portals
- Agent: morning load risk board
- What AI does: Sorts uncovered and at-risk loads
- Helps: Focused dispatch day
- Control: Confirm status
- What AI does: Feasible drivers with deadhead and HOS fit
- Helps: Faster, safer assignments
- Control: Dispatcher confirms hours
- What AI does: Demand vs. capacity by day and region
- Helps: Secure capacity early
- Control: Update daily
- What AI does: Orders stops around windows
- Helps: Less detention
- Control: Use truck-routing software
- What AI does: Drafts delay and ETA messages
- Helps: Proactive service
- Control: Verify ETA
- What AI does: Weather and closure playbooks
- Helps: Safe, organized response
- Control: Drivers can always stop
Dispatch Implementation Checklist
WorkflowBefore You Start
- Export load board fields
- Confirm telematics connector permissions
- Write dispatch and HOS rules
- Baseline: on-time %, deadhead
- Name a pilot dispatcher
After You Start
- Review options before assigning
- Track on-time and deadhead
- Audit HOS conflicts flagged
- Update customer templates
- Review connector access quarterly
- Safety: Never plan loads that require exceeding hours-of-service limits; dispatchers never pressure drivers.
- Assignments: People assign loads and confirm with drivers.
- Action-capable connectors: Motive MCP can message drivers and create dispatches; enable actions only with explicit owner approval.
- Routing: Use truck-legal routing tools for miles and routes.
- Accuracy: Confirm ETAs before telling customers.
AI for Carrier Sourcing & Vetting
Use AI to shortlist carriers, benchmark rates, and flag fraud indicators. Compliance approves carriers using authoritative sources.
- Prompts: lane carrier shortlist, fraud red flag review, insurance certificate review
- Skills: carrier sourcing & procurement; carrier vetting & fraud screening
- Connect: TMS carrier history, DAT or Truckstop exports, vetting tool reports
- Agent: lane capacity finder; carrier fraud red-flag screener
- What AI does: Ranks carriers by your lane history
- Helps: Faster coverage
- Control: Vetting first
- What AI does: Summarizes spot rates and trends
- Helps: Aligned buy and sell
- Control: Pricing decides
- What AI does: Award scenarios for RFPs
- Helps: Clear trade-offs
- Control: Procurement awards
- What AI does: Flags identity and double-brokering signs
- Helps: Stop fraud before tender
- Control: Verify via FMCSA and known contacts
- What AI does: Checks coverage against requirements
- Helps: Catch gaps
- Control: Confirm with agent
- What AI does: Summarizes authority and insurance changes
- Helps: Act on alerts
- Control: Compliance suspends
Carrier Vetting Control Checklist
ControlsBefore You Start
- Write vetting requirements
- Set up carrier monitoring
- Document payment change verification
- Train reps on red flags
- Baseline: fraud incidents, time to cover
After You Start
- Vet every carrier before every tender
- Review red-flag packets daily
- Verify payment changes by phone
- Track fraud attempts
- Update requirements quarterly
- Approval: Compliance approves carriers; AI only flags.
- Sources: Use FMCSA records, insurer confirmation, and your vetting tool, never details supplied in a suspicious request.
- Payments: Verify any banking or factoring change by calling a known number; two-person approval.
- Fairness: Apply the same requirements to every carrier.
- Incidents: Escalate suspected fraud to leadership, insurer, and law enforcement as appropriate.
AI for Freight Billing, Audit & Accessorials
Use AI to get loads billed, audit invoices, and capture detention. Billing and AP approve every invoice and payment.
- Prompts: billing readiness, carrier invoice audit, detention package
- Skills: freight billing & invoice audit; detention & accessorial management
- Connect: TMS billing data, document management, ELD timestamps
- Agent: billing readiness monitor; carrier invoice auditor; detention capture assistant
- What AI does: Lists loads ready and missing documents
- Helps: Faster cash
- Control: Customer rules
- What AI does: Matches invoices to rate confirmations
- Helps: Prevent overpayment
- Control: AP decides
- What AI does: Catches carrier and factor double-billing
- Helps: Stop losses
- Control: Confirm with both
- What AI does: Builds evidence-backed detention bills
- Helps: Collect what you're owed
- Control: Contract terms
- What AI does: Accessorials paid but not billed
- Helps: Protect margin
- Control: Check rebillability
- What AI does: Evidence-based customer replies
- Helps: Faster resolution
- Control: Manager decides credits
Billing & Audit Control Checklist
ControlsBefore You Start
- Collect customer billing requirements
- Document accessorial approval workflow
- Confirm ELD timestamp access
- Baseline: days to bill, DSO
- Define duplicate rules
After You Start
- Audit every carrier invoice
- Bill detention weekly
- Track leakage monthly
- Review disputes
- Update rules with contract changes
- Payments: AI never approves or pays invoices.
- Payment changes: Verify banking changes independently.
- Evidence: Bill accessorials only with documented proof.
- Contracts: Apply each customer's terms exactly.
- Credits: Billing managers approve credits and short-pays.
AI for DOT & Hours-of-Service Compliance
Use AI to find HOS patterns, track DQ files, and prepare for audits. Safety and compliance staff make every determination.
- Prompts: HOS violation trend review, DQ file audit, compliance calendar
- Skills: HOS & ELD compliance review; driver qualification & DOT readiness; fleet safety & maintenance
- Connect: Samsara MCP (read-only HOS and DVIR data) or Motive MCP (fleet admin only), compliance software
- Agent: HOS pattern watch; DQ file & compliance calendar monitor; PM & defect tracker
- What AI does: Violations by type, terminal, and dispatcher
- Helps: Fix planning causes
- Control: Compliance reviews
- What AI does: Matches events to likely drivers for review
- Helps: Clean logs
- Control: Admin verifies
- What AI does: Missing and expiring items
- Helps: Audit readiness
- Control: Compliance decides dispatchability
- What AI does: Filings, renewals, and program deadlines
- Helps: On-time filings
- Control: Confirm requirements
- What AI does: Trends and corrective actions
- Helps: Lower scores
- Control: DataQs reviewed by staff
- What AI does: Overdue PM and open DVIR defects
- Helps: Fewer OOS violations
- Control: Shop decides
DOT Compliance Checklist
ControlsBefore You Start
- Map compliance obligations by operation type
- Confirm telematics data access and roles
- Keep drug and alcohol data out of AI
- Baseline: violations, findings
- Name compliance owners
After You Start
- Review HOS patterns weekly
- Audit DQ files monthly
- Track filings
- Coach based on patterns
- Self-audit annually
- Authority: FMCSA regulations and your compliance staff decide; AI summaries can be wrong.
- Records: Never edit ELD logs or create records after the fact.
- Drug & alcohol: Results stay with the DER and out of AI tools.
- Dispatch: Never pressure drivers; HOS conflicts go to compliance.
- Connectors: Samsara MCP is read-only; Motive MCP can take actions and shows fleet-wide data to connected admins.
AI for Cargo Claims
Use AI to organize claim files, track deadlines, and find root causes. Liability and settlement decisions stay with claims managers and counsel.
- Prompts: claim intake summary, carrier claim notice, claims root-cause analysis
- Skills: cargo claims management
- Connect: claims system or shared folders, TMS documents, temperature logs
- Agent: claims deadline tracker
- What AI does: Summarizes claims and missing documents
- Helps: Organized files
- Control: Manager reviews
- What AI does: Tracks filing and response deadlines
- Helps: Protect rights
- Control: Confirm dates
- What AI does: Drafts formal claim notices
- Helps: Consistent language
- Control: Counsel for large claims
- What AI does: Organizes reefer data
- Helps: Faster evaluation
- Control: Experts determine impact
- What AI does: Claims by cause, carrier, and lane
- Helps: Prevent repeat claims
- Control: Standard categories
- What AI does: Communicates decisions clearly
- Helps: Fewer escalations
- Control: Manager decides
Claims Implementation Checklist
ControlsBefore You Start
- List deadlines that apply (contracts, 49 CFR Part 370 for carrier claims)
- Organize document access
- Standardize cause codes
- Baseline: cycle time
- Clarify insurer reporting
After You Start
- Review deadlines daily
- Close documentation gaps
- Report root causes quarterly
- Track recovery
- Share prevention guides
- Liability: Claims managers and counsel decide; AI organizes.
- Admissions: Drafts never admit liability.
- Deadlines: Confirm every deadline against contracts and law.
- Evidence: Never alter documents or photos.
- Insurance: Follow policy reporting requirements.
AI for Customer Quotes & Pricing
Use AI to build quotes, review contract rates, and analyze wins and losses. Pricing leaders approve rates.
- Prompts: spot quote builder, contract rate review, quote win/loss
- Skills: customer quoting & pricing; customer account management
- Connect: market rate tools, TMS lane history, CRM
- Agent: spot quote assistant; customer margin & AR watch
- What AI does: Buy and sell ranges with margin
- Helps: Fast, consistent quotes
- Control: Rep confirms
- What AI does: Rates vs. cost and market
- Helps: Fix losing lanes
- Control: Check terms
- What AI does: Win rate by lane and price position
- Helps: Smarter pricing
- Control: Ask customers
- What AI does: Data-based lane pricing
- Helps: Profitable bids
- Control: Leader approves
- What AI does: Clear customer-facing terms
- Helps: Fewer disputes
- Control: Pricing approves
- What AI does: Performance reviews for customers
- Helps: Retention
- Control: Ops verifies data
Pricing Implementation Checklist
WorkflowBefore You Start
- Set margin targets
- Export lane history
- Set up market rate data access
- Document accessorial schedule
- Baseline: win rate, margin
After You Start
- Track quote turnaround
- Review contract lanes monthly
- Analyze win/loss quarterly
- Hold QBRs
- Update fuel tables
- Approvals: Pricing leaders approve rates and bids.
- Commitments: Only promise service you can deliver.
- Confidentiality: Customer rates stay confidential.
- Accuracy: Quotes include clear validity and inclusions.
- Markets: Benchmarks are averages; judgment decides.
Set Up & Get Running
If you already use prompts, here is how to turn one into a repeatable skill and then a useful first agent. Start with a task that already works.
From prompt to first agent
- Choose a daily or weekly task
- Use real data
- Good first choices: carrier invoice audit, detention package, or load board review
- Write down inputs, output, and reviewer
- Test on three real weeks
- Improve when results are unclear
- Start with TMS exports and read-only telematics
- Keep action-capable connectors off
- Keep restricted driver data out
- Start with an audit or monitoring task
- Set when it runs; a person reviews
- Track dollars recovered and time saved
It's useful, measurable, and easy to check. It compares invoices to rate confirmations and flags duplicates and unapproved charges. It doesn't approve, pay, or change payment details.
Copy these builders
Use the first after a prompt works. Use the second only after the skill has been tested on real work.
Days 1-30: Foundation
- Name an AI lead in operations
- Write data, safety, and approval rules
- Set up business accounts
- Connect one read-only source (e.g., Samsara MCP or TMS exports)
- Pick 1 pilot (invoice audit or detention)
- Record baseline
Days 31-60: Expand
- Save the pilot as a skill
- Add a 2nd workflow (load board review or fraud screening)
- Draft the AI use policy
- Train dispatch and carrier sales
- Measure recovered dollars and time
Days 61-90: Standardize
- Add a 3rd workflow (HOS patterns or claims deadlines)
- Adopt policy
- Pilot one review-only agent
- Write SOPs
- Report results
- Plan next wave
Implementation Success Measures
Measurement30-Day Targets
- Pilot running
- Baseline recorded
- Read-only access only
- No restricted data used
60-Day Targets
- Dollars recovered or errors caught tracked
- Second workflow live
- Policy drafted
- Teams trained
90-Day Targets
- 3 workflows with SOPs
- Policy adopted
- First agent piloted
- Measurable impact
- Decision on next wave
How the Pieces Fit Together
Start with a prompt. Save the ones that work. Add the systems and files they need. Then use an agent for recurring work, with someone checking the result.
- Ask AI to do one clear task
- Use it for a quick draft, analysis, or explanation
- Check the result before you use it
- Save the step-by-step instructions for a task your team repeats
- Add your policies, examples, and review steps
- Update it when your process or regulations change
- Bring the right systems, files, and reports into the task
- Stop copying and pasting information by hand
- Start read-only, with only the sources the task needs
- Run a skill on a schedule or when something happens
- Gather the information, prepare the work, and flag issues
- Leave assignments, approvals, payments, and safety decisions with people
Where AI Can Help Logistics Teams
Use AI to get routine logistics work into a clear first draft. Keep safety, carrier approvals, payments, and pricing with your team.
- Risk-sorted load boards
- Feasible driver and carrier options
- Capacity plans
- Customer update drafts
- Billing readiness and invoice audits
- Detention capture
- Accessorial leakage
- Margin and AR reviews
- Fraud red-flag screening
- HOS and DQ tracking
- PM and defect lists
- Claim files and deadlines
- Assignments and tenders
- Carrier approvals
- Compliance determinations
- Payments, claims, and pricing
AI Capabilities Explained
No jargon. What each kind of AI does, and what it looks like in logistics.
Claude, ChatGPT, Gemini, and Copilot write, summarize, and follow instructions.
In logistics: quote emails, claim notices, SOPs, shift handoffs.
AI that reads your systems through connectors.
In logistics: asking Claude about HOS status through Samsara MCP.
Connectors that can change records or send messages.
In logistics: Motive MCP can message drivers or create dispatches; use only with approval.
Compares records to rules and flags exceptions.
In logistics: invoices vs. rate confirmations, DQ files vs. requirements.
Reads fields from scanned documents.
In logistics: BOL and POD data, with human review of flagged fields.
Finds trends and anomalies in data.
In logistics: dwell by facility, HOS patterns by terminal, fraud signals.
Guardrails: Safety, Fraud & Compliance
Simple rules for using AI in logistics. Drivers' safety, customers' freight, and carriers' payments depend on getting these right.
- Drug and alcohol results and medical details
- Driver SSNs and license numbers
- Carrier banking details
- Customer contracts beyond what the task needs
- Never plan around HOS limits
- Dispatch never pressures drivers
- Units with safety defects stay parked
- Drivers can always stop when unsafe
- Vet every carrier before every tender
- Verify payment changes by calling known numbers
- AI flags; compliance decides
- Escalate suspected fraud fast
- Start read-only (e.g., Samsara MCP)
- Action-capable connectors only with approval
- Review who connected and what they can see
- Review quarterly
- Never edit ELD logs via AI
- Never alter BOLs, PODs, or photos
- Bill only with evidence
- Keep an audit trail
- Load assignments and tenders
- Carrier approvals and suspensions
- Compliance determinations and coaching
- Claims, payments, and pricing
Logistics AI Readiness Checklist
ControlsLeadership & Policy
- Executive sponsor named
- AI use policy adopted
- Business accounts with training off
- Connector permission rules written
- Dispatch, carrier, billing, and safety teams trained
Practice & Review
- Read-only connectors by default
- Human approval on tenders, payments, and pricing
- Fraud verification procedures enforced
- Restricted data kept out
- Quarterly review of AI use and incidents
Scope: Dispatch, carrier sales, pricing, billing, AP and AR, safety, compliance, claims, and leadership.
Approved tools: [e.g., Claude Team; Samsara MCP (read-only); Microsoft Copilot; TMS exports to a shared folder]
Approved uses: load board triage, option lists, quote ranges, invoice audits, detention packages, HOS pattern reviews, DQ tracking, claim file organization, reports.
Never enter into AI tools: drug and alcohol results, medical details, driver SSNs or license numbers, carrier banking details.
Human decisions: load assignments, tenders, carrier approvals, HOS and compliance determinations, coaching, claims liability, payments and payment changes, pricing.
Connectors: Read-only by default. Action-capable connectors (e.g., Motive MCP) only for specific approved actions with logging.
Safety: No AI output may be used to pressure a driver or plan beyond hours-of-service limits.
Approval: [COO / Safety Director / CFO], [date]
AI Maturity Model for Logistics
See where your operation is today. Choose a realistic next step.
Curious + Ad Hoc
- Individuals use AI for emails
- No data or safety rules
- No connectors
- No shared prompts
Shared Prompts & Rules
- Data, safety, and approval rules
- Business accounts
- Prompt library
- One read-only connection
Embedded Workflows
- Skills for audits, detention, dispatch, and compliance
- Adopted policy
- Regular TMS exports
- Measured results
Connected & Measured
- Review-only agents with owners
- Governed action-capable connectors
- Fraud and safety metrics tracked
- Quarterly reviews
Maturity Self-Assessment
AssessmentLeadership
- Sponsor
- Policy
- Training
- Connector rules
Practice
- Prompt library
- Skills
- SOPs
- Exports
Protection
- HOS rules
- Fraud verification
- Restricted data out
- Human approvals
Measurement
- Baseline
- Dollars recovered
- Errors caught
- Quarterly review