.jpg)
The integration challenge
Enterprise supply chains run on dozens of systems: ERPs, TMSs, WMSs, OMSs, planning tools, BI platforms, and increasingly, AI assistants and copilots. Each system has its own view of the supply chain. Connecting them has traditionally required point-to-point integrations: custom APIs, middleware, and months of development per connection. As enterprises deploy AI assistants (Copilot for M365, internal GPT-based tools, custom agents), each one needs access to supply chain data and actions.
The MCP Gateway architecture
FourKites has built a Model Context Protocol (MCP) gateway that exposes FourKites data and actions as tools that any MCP-compatible AI system can call. MCP is an emerging standard for connecting AI systems to external data sources and action endpoints. The FourKites MCP gateway means that any AI assistant in the enterprise can ask 'What is the ETA for my shipment?' or 'Reschedule the appointment at the Chicago DC' and receive a response grounded in live FourKites data.
Available tool categories
The gateway exposes tools across four categories. Query tools provide read access to shipment status, ETAs, order tracking, facility status, and carrier performance. Action tools enable appointment scheduling, carrier communication, exception escalation, and workflow triggering. Analytics tools provide FourSight-style natural language analytics. Configuration tools enable workflow management and agent configuration.
Security and scoping
Every MCP call is authenticated, scoped to the requesting user's permissions, and logged. A customer's AI assistant can only access that customer's data. Actions require the same authorization levels as the FourKites UI. The gateway enforces the same business rules (escalation thresholds, approval requirements, notification preferences) that govern agent behavior in Loft.
Why this matters
The MCP gateway transforms FourKites from an application into infrastructure. Instead of being one of many tabs in a supply chain professional's browser, FourKites becomes a capability that any AI system in the enterprise can invoke. The supply chain intelligence that powers Tracy, Alan, Cassie, and the entire Digital Workforce becomes available to whatever AI interface the enterprise adopts. FourKites does not need to be the interface. It needs to be the intelligence.
Newsletter
Stay Informed. Join 30,000+ monthly readers and get exclusive ebooks, reports, and industry insights from FourKites every week.
Thank you for your submission!
Oops! Something went wrong while submitting the form.
One billion hours of operational work completed by AI agents over the next decade.
The supply chains that adopt autonomous execution in the next 24 months will define the competitive standard for the next decade. The ones that do not will spend that decade trying to catch up.
Ready? Talk to an Outcome Advisor
