--- title: "MaintainX Integration" description: "Connect AI agents to MaintainX, the work order and maintenance management platform. Agents open work orders when equipment faults are detected, update and assign them to technicians, manage assets and locations, and read parts inventory and team members." url: https://flowrunner.ai/integrations/maintainx date_modified: 2026-08-01T02:40:32-07:00 --- # MaintainX [ERP](https://flowrunner.ai/integrations/category/finance-accounting) Connect AI agents to MaintainX, the work order and maintenance management platform. Agents open work orders when equipment faults are detected, update and assign them to technicians, manage assets and locations, and read parts inventory and team members. [Verified](https://flowrunner.ai/integrations/verified "What does verified mean?") · 10 actions · API key · available [MaintainX website](https://getmaintainx.com/) · [Platform Documentation](https://api.getmaintainx.com/v1/docs/) · Capability data verified 2026-07-28 1. A monitoring system reports a fault on a packaging line compressor 2. Get Asset pulls the equipment's serial number, model, and location for context 3. List Work Orders checks whether an open order already covers this asset and fault 4. List Users resolves the technicians on shift for that location 5. Create Work Order opens the job with High priority, the asset linked, and the on-shift technician assigned 6. The maintenance channel gets the work order link, asset detail, and due date 7. The maintenance supervisor confirms any escalation to Urgent before an after-hours callout goes out 8. The agent tracks the order's status and re-alerts if it sits Open past its response window ## What This Integration Enables Maintenance management fails at the seams between detection and dispatch: the machine knew it was failing before anyone wrote it down. MaintainX holds the work orders, assets, locations, parts, and people; this connector lets FlowRunner agents operate that system at the speed of the events it exists to track. Fault alerts become work orders with the asset linked and a technician assigned. Status changes propagate to the channels where teams live. The backlog becomes a report that arrives instead of a discovery that surprises. This is the connector where FlowRunner's andon-cord philosophy is most literal. In lean manufacturing, anyone can stop the line when something is wrong. Here, any system can pull the cord: the sensor, the ERP, the incident tracker, and the workflow ensures a human technician is dispatched with full context. - Equipment alerts open deduplicated work orders with priority mapped from severity - Assets and locations sync in from other systems, so the CMMS reflects the real plant - Assignment logic reads the team roster and the location, not a hardcoded name - Backlog, aging, and completion reporting publish on schedules to sheets and chat - Parts inventory stays visible to the workflows deciding whether a repair can start today What agents deliberately do not do here: declare work finished or wake up the night crew on their own authority. Both run through FlowRunner's [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) layer. ### Without FlowRunner **Faults travel by hallway**: The operator notices, tells a lead, who tells maintenance, who opens a ticket, and the timestamp lies about all of it **Duplicate work orders collide**: Two people report the same failure, two orders open, and two technicians walk to the same machine **The backlog is invisible until it hurts**: Open orders accumulate quietly, and leadership learns about the maintenance debt from the downtime it causes ### With FlowRunner **Faults open their own work orders**: Alerts become assigned, prioritized, asset-linked orders in seconds, deduplicated against what is already open **Assignment follows the roster**: Agents resolve who is on shift at which location and assign accordingly, not to whoever is memorable **The backlog reports itself**: Open and in-progress orders publish to dashboards and channels on a schedule, aged and prioritized ## Use Case Scenarios ### The incident that dispatched its own technician A [ServiceNow](https://flowrunner.ai/integrations/servicenow) incident fires for a failing HVAC unit. The agent maps the incident severity to a MaintainX priority, pulls the asset record with Get Asset, checks List Work Orders for an existing open order on that unit, and finding none, calls Create Work Order with the incident description and the on-shift technician assigned. The work order link posts back to [Slack](https://flowrunner.ai/integrations/slack) and onto the incident. Detection to dispatch took less time than the old process took to find the right person to tell. ### The maintenance backlog that briefs the morning meeting On a schedule, the agent walks List Work Orders filtered by status, ages each against its due date, and writes the picture to [Google Sheets](https://flowrunner.ai/integrations/google-sheets): open orders by location, in-progress by technician, overdue flagged. High-priority items landing overnight get called out in [Microsoft Teams](https://flowrunner.ai/integrations/microsoft-teams) with asset and due date. The morning meeting starts from the same live sheet every day, and "how big is the backlog" stops being a question that requires someone to go count. ### New equipment that enters the system before it enters service Procurement closes a purchase for three new mixers. The agent creates each in MaintainX with Create Asset, name, description, location assigned from List Locations, as soon as the delivery confirms. Commissioning work orders open automatically, assigned to the installation crew. When the first fault eventually comes, the asset record with serial and model detail is already there, because the asset existed in the CMMS from the day it existed in the building. ## Human-in-Loop Highlight Two operations on this connector stay human. The first is Update Work Order setting status to Done: a maintenance record is an attestation that someone inspected the machine and the work is complete, and an agent marking orders Done on a schedule or a signal would quietly convert the compliance trail into fiction. FlowRunner agents move orders to In Progress or On Hold and re-alert on stalls, but Done is entered by the technician who stood at the equipment. The second is escalation to Urgent via Create Work Order or Update Work Order, because Urgent is not a label, it is an event: it pages people, authorizes after-hours labor cost, and can stop production. The agent assembles the case, fault data, asset history, open-order context, and the maintenance supervisor makes the call. Everything between those two edges runs at machine speed. Agent processes routinely Detects exception requiring judgment Clear match Continues automatically Ambiguous Routes to human via preferred channel Human decides Agent resumes with decision ## Agent Capabilities 10 actions ### Work Orders 4 - **List Work Orders** Lists work orders in the organization, most recent first. Supports filtering by status (Open, In Progress, On Hold, Done) and by the ID of an assigned user, and cursor-based pagination. Returns up to 200 work orders per page (default 50). Use the returned cursor to fetch the next page. - **Get Work Order** Retrieves a single work order by its ID, including its title, description, status, priority, linked asset and location, assignees, due date, and timestamps. - **Create Work Order** Creates a new work order. A title is required; you can also set a description, priority (Low, Medium, High, Urgent), the asset the work is performed on, and a list of assignee user IDs. Returns the created work order with its assigned ID. - **Update Work Order** Updates fields on an existing work order. Only the fields you provide are changed; omitted fields are left as-is. You can change the title, description, priority, status, linked asset or location, assignees, and due date. ### Assets 3 - **List Assets** Lists assets (equipment and machinery) in the organization with cursor-based pagination. Optionally filter by the location the assets belong to. Returns up to 200 assets per page (default 50). Use the returned cursor to fetch the next page. - **Get Asset** Retrieves a single asset by its ID, including its name, description, location, serial number, and manufacturer/model details. - **Create Asset** Creates a new asset. A name is required; you can also set a description and assign it to a location. Returns the created asset with its assigned ID. ### Locations 1 - **List Locations** Lists locations (sites, buildings, and areas) in the organization with cursor-based pagination. Returns up to 200 locations per page (default 50). Use the returned cursor to fetch the next page. ### Parts 1 - **List Parts** Lists parts and inventory items in the organization with cursor-based pagination. Returns up to 200 parts per page (default 50). Use the returned cursor to fetch the next page. ### Users 1 - **List Users** Lists users (team members) in the organization with cursor-based pagination. Use the returned user IDs when assigning work orders. Returns up to 200 users per page (default 50). ## Frequently Asked Questions ### What can FlowRunner do with MaintainX? FlowRunner agents can run List Work Orders, Get Work Order, and Create Work Order in MaintainX, plus 7 more actions. ### Does connecting MaintainX to FlowRunner require OAuth? No. MaintainX connects to FlowRunner with an API key, no OAuth flow required. ### Can MaintainX trigger a FlowRunner workflow automatically? MaintainX doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger. **Work at MaintainX?** This integration exposes MaintainX to AI agents on every FlowRunner plan, including through MCP, at no cost to you. [See what FlowRunner offers integration partners](https://flowrunner.ai/integrations/partners), including how to keep this page current. --- Markdown version of https://flowrunner.ai/integrations/maintainx. Site index: https://flowrunner.ai/llms.txt