Onfleet MCP. Manage tasks and track every driver from chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Onfleet MCP Server connects your entire delivery operation to any AI agent. You can create, update, and delete delivery tasks, track drivers in real time, check exact ETAs, and pull historical task reports—all from a single conversation.
What your AI agents can do
Complete task override
Force-completes any delivery task immediately in the system.
Create delivery task
Builds and sends a brand new delivery job to Onfleet.
Delete delivery task
Archives or removes an existing delivery task record.
Start new deliveries using create_delivery_task and adjust details or force completion via update_delivery_task.
List all active workers (list_fleet_workers) and check their current capacity and schedule to see who's available for a new route.
Retrieve granular details about any specific delivery job using get_task_details.
Pull a list of tasks that occurred within a specific date range using list_tasks_by_date.
List all dispatch hubs (list_dispatch_hubs) or view your entire team structure (list_fleet_teams).
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Onfleet MCP Server: 10 Tools for Delivery Operations
Use these ten tools to manage the entire task lifecycle—from initial creation and scheduling to live tracking, auditing, and final completion.
019d75e6complete task override
Force-completes any delivery task immediately in the system.
019d75e6create delivery task
Builds and sends a brand new delivery job to Onfleet.
019d75e6delete delivery task
Archives or removes an existing delivery task record.
019d75e6get task details
Retrieves all specific information, status, and ETA for one single delivery task.
019d75e6get worker schedule
Pulls the detailed shift times and availability window for a specific driver.
019d75e6list dispatch hubs
Lists all operational dispatch hubs, including their geographic coordinates.
019d75e6list fleet teams
Retrieves the names and details of every delivery team in your network.
019d75e6list fleet workers
Lists all active drivers or workers, showing their current status and capacity.
019d75e6list tasks by date
Finds all delivery tasks that occurred within a specific start and end date range.
019d75e6update delivery task
Changes any field—like address, status, or recipient—on an existing task.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Onfleet, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
Onfleet connects your whole delivery operation right into your AI agent. You run complex logistics tasks using natural language; you don't have to jump between dashboards or open a dozen separate apps just to manage last-mile deliveries.
How Onfleet MCP Works
- 1 Subscribe to the Onfleet server and provide your API key.
- 2 Tell your AI client exactly what you need (e.g., 'List all drivers online with capacity').
- 3 The agent executes the necessary tool calls, pulls the live data, and gives you a direct answer in chat.
The bottom line is that your AI client talks directly to Onfleet's backend, giving you real-time operational data without needing a web browser.
Who Is Onfleet MCP For?
Logistics managers who hate switching between five different dashboards. Ops teams constantly clearing exceptions and fixing stuck orders on the fly. Customer support agents who need instant, accurate ETAs for angry clients.
Checks fleet capacity using list_fleet_workers and assigns tasks to optimize daily routes.
Forces-completes or deletes a task that failed in the system, then uses update_delivery_task to correct the record.
Looks up a delivery ID using get_task_details to give a customer an accurate, real-time ETA for their package.
What Changes When You Connect
- Stop switching tabs to check status. Use
get_task_detailsto pull a task's current ETA instantly, giving you the information in one conversational turn. - Need an audit? Instead of running reports manually, use
list_tasks_by_date. You tell your agent the dates, and it gives you a summary of all deliveries that happened then. - Fix exceptions on the fly. If a task is stuck or needs new details, don't log into the dashboard; just call
update_delivery_taskvia your AI client. - Know who can take the next job. Check real-time capacity using
list_fleet_workers. You immediately see which drivers are online and how many active tasks they have. - Keep track of personnel. Use
get_worker_scheduleto get a worker's exact shift times, ensuring you don't assign them outside their paid window.
Real-World Use Cases
The Dispatch Manager needs an immediate headcount.
A manager sees three high-priority tasks come in simultaneously. Instead of checking the dashboard to see who has space, they ask their agent: 'Which drivers are online and what is their available capacity?' The agent uses list_fleet_workers and provides a list of available drivers immediately.
Customer Support needs an urgent ETA.
A customer calls asking where their package from yesterday is. Instead of making the customer wait while you check multiple systems, you ask your agent to run get_task_details for that tracking ID. You get the live status and accurate ETA in seconds.
The Ops Team needs to clear a failed job.
A driver reports an address was wrong, so the task is stuck and pending. The ops team member tells their agent: 'Update task XYZ; the address is actually 45 Elm St.' The agent uses update_delivery_task to correct the record instantly.
Auditing yesterday's performance.
It's the end of the month, and you need a report on all deliveries. You prompt your agent: 'Show me all tasks from October 1st through October 30th.' The agent runs list_tasks_by_date and hands back a categorized list.
The Tradeoffs
Checking every task manually.
A user tries to see if Task A's status changed by first calling list_tasks_by_date for the whole day, then scrolling through 50 results to find Task A. It’s slow and noisy.
→
You don't need to list everything. Just use get_task_details and give it the task ID directly. You get only the information you asked for.
Assuming status changes are automatic.
A user thinks a failed delivery just 'fixes itself' and tries to track it, but no data comes back because they didn't change the core record.
→
If you need to fix a task that got stuck or needs new details, don’t wait. Use update_delivery_task with the correct information.
Asking for status without IDs.
A user asks: 'What's happening with the deliveries?' The agent can't answer because it needs a time window or specific ID to run any tool effectively.
→
Be precise. To see everything that happened yesterday, use list_tasks_by_date. If you only care about drivers, start by running list_fleet_workers.
When It Fits, When It Doesn't
Use this server if your core business loop involves constantly managing the state of moving assets. You need to know who's available right now (list_fleet_workers), what status a specific job is in (get_task_details), and you have exceptions that require manual correction (update_delivery_task).
Don't use this if your process is simple, one-way data entry (e.g., just logging bookings). For basic scheduling validation without live tracking needs, a dedicated calendar or simpler booking tool might be enough. If you only need to look up static addresses and never care about real-time status or fleet capacity, you don't need the full power of Onfleet.
If your workflow involves anything that requires 'Where is it?' or 'Who has time?', this server gives you the tools.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Onfleet. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 10 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Checking delivery status shouldn't require three different dashboards.
Today, checking one delivery means bouncing between your main dispatch board, the driver communication app for location pings, and then opening a separate analytics report just to see if it was completed. You spend five minutes copying IDs, switching tabs, and hoping nothing breaks.
With this MCP server, you skip all that noise. Tell your agent which task ID you're looking at. The agent runs `get_task_details` and gives you the current status, ETA, and history—all in one chat bubble. It’s immediate.
Onfleet MCP Server: Control tasks and schedules with precision.
Forget calling the operations desk every time a task fails or needs an address update. You don't need to find a form, log in, click through six screens, and submit the fix. The process is too slow for real-time logistics.
Now you just tell your agent, 'Update Task XYZ with this new detail.' The server runs `update_delivery_task` directly. It’s direct. It's fast. It works.
Common Questions About Onfleet MCP
How do I check the status of a delivery using Onfleet MCP Server? +
You use get_task_details. Give the agent the task ID, and it pulls the current location, ETA, and full status. You'll know exactly where that package is.
Can I see who is available using Onfleet MCP Server? +
Yes, you can list all drivers and check their capacity with list_fleet_workers. This shows who is online and how many tasks they are currently handling.
How do I fix a task that failed yesterday using Onfleet MCP Server? +
You use update_delivery_task or complete_task_override. You give the agent the necessary details, and it handles the correction for you.
What if I need an audit of tasks from last week? Use Onfleet MCP Server. +
Run list_tasks_by_date and specify your start and end dates. This pulls a comprehensive list of every task that ran in that period.
How do I get a specific driver's shift hours using Onfleet MCP Server? +
You use the get_worker_schedule tool, providing the worker ID or name. This returns their exact start and end times, detailing their availability windows for that day.
How do I see all available delivery teams using Onfleet MCP Server? +
Call the list_fleet_teams tool to retrieve every defined team. This shows you the full organizational structure, regardless of current task assignments or active drivers.
What is the process for modifying an existing delivery task using Onfleet MCP Server? +
You use the update_delivery_task tool to change details like addresses or recipients. You must provide the unique Task ID so we only modify the correct record.
How do I ensure my AI client is connected and authenticated with Onfleet MCP Server? +
Authentication requires passing your valid Onfleet API Key during setup. Your agent uses this key to confirm access, ensuring all commands run on your authorized account.
Can my AI agent create a complete delivery task with address and recipient? +
Yes. Pass a JSON object with destination address, recipient name, phone, and any notes — the agent calls the Onfleet Tasks API and returns the created task with its tracking ID. You can also update or delete tasks the same way.
How does force-completing a task differ from a normal driver completion? +
Force-completing bypasses the driver app flow — useful when a delivery was made but the driver forgot to mark it done, or for admin overrides. The agent sends the completion details (success/failure and notes) directly to the Onfleet API, updating the task status without driver interaction.
Can I see which drivers are online right now and their schedules? +
Yes. List all workers to see their online/offline status, active task count, and current location. Then query any specific worker's schedule to see their shift times and availability windows. Perfect for capacity planning and real-time dispatch decisions.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Dashdoc
Equip your AI agent to manage transport orders, fleet, and delivery addresses via the Dashdoc API.
Odoo Manufacturing
Create manufacturing orders, manage BOMs, work orders, and work centers — Odoo MRP through natural conversation.
Uniconta
Automate ERP workflows via Uniconta — retrieve debtors, query invoices, manage GL accounts, and list inventory directly from any AI agent.
You might also like
Google Calendar
Sync and orchestrate your agenda securely — scan, schedule, and manipulate Google Calendar events natively in chat.
Focus Admin
Send CC-e, inutilize numbering, distribute DF-e and list fiscal documents via Focus NFe API.
HID Origo
Automate identity and access management via HID Origo — manage mobile identities, users, and readers directly from any AI agent.