Freight MCP
A freight MCP server is a Model Context Protocol implementation that exposes freight network operations as tools an AI agent can call. Quote, book, track, retrieve documents — each as a structured MCP tool with typed inputs and outputs. The agent does not learn the freight API by reading documentation; the MCP server tells the agent what it can do, then handles the API call when the agent invokes a tool. This lets Claude, Cursor, ChatGPT, or any MCP-compatible AI book real freight without a custom integration.
Why it matters
AI agents have been able to chat about freight for years but could not actually move it. The MCP layer collapses that gap. An agent that runs shell commands, edits code, and now books freight is a different operations layer than a chatbot. For shippers running ops through AI assistants, an MCP-equipped freight network means the agent completes a quote-to-booking workflow in seconds without copy-paste from a portal. The shape of freight operations changes from human-in-the-loop to agent-in-the-loop.
When to use it
Plug a freight MCP server into any AI assistant your team uses for operations — Claude Desktop, Cursor, Claude Code, or ChatGPT desktop. The MCP becomes a tool the agent reaches for when a freight task comes up. Most useful when the freight workflow is part of a larger task (sourcing inventory, scheduling production, recovering a missed shipment) rather than a standalone booking session. Less useful for one-off shipments where the portal flow is faster.
How Warp thinks about it
Warp ships the first MCP server for booking real freight — warp-agent-mcp on npm, 17 tools covering quote/book/track/document/cancel across LTL, FTL, box truck, cargo van, and multi-stop FTL. `npx warp-agent-mcp` connects to any MCP-compatible client. The same Bearer-token key works for the REST API, the MCP server, and the CLI agent — one auth flow, three surfaces. No demo, no sales call, no separate integration project.
Frequently asked questions about freight mcp
What is freight mcp?
A freight MCP server is a Model Context Protocol implementation that exposes freight network operations as tools an AI agent can call. Quote, book, track, retrieve documents — each as a structured MCP tool with typed inputs and outputs. The agent does not learn the freight API by reading documentation; the MCP server tells the agent what it can do, then handles the API call when the agent invokes a tool. This lets Claude, Cursor, ChatGPT, or any MCP-compatible AI book real freight without a custom integration.
Why does freight mcp matter in freight?
AI agents have been able to chat about freight for years but could not actually move it. The MCP layer collapses that gap. An agent that runs shell commands, edits code, and now books freight is a different operations layer than a chatbot. For shippers running ops through AI assistants, an MCP-equipped freight network means the agent completes a quote-to-booking workflow in seconds without copy-paste from a portal. The shape of freight operations changes from human-in-the-loop to agent-in-the-loop.
When should you use freight mcp?
Plug a freight MCP server into any AI assistant your team uses for operations — Claude Desktop, Cursor, Claude Code, or ChatGPT desktop. The MCP becomes a tool the agent reaches for when a freight task comes up. Most useful when the freight workflow is part of a larger task (sourcing inventory, scheduling production, recovering a missed shipment) rather than a standalone booking session. Less useful for one-off shipments where the portal flow is faster.
How does Warp handle freight mcp?
Warp ships the first MCP server for booking real freight — warp-agent-mcp on npm, 17 tools covering quote/book/track/document/cancel across LTL, FTL, box truck, cargo van, and multi-stop FTL. `npx warp-agent-mcp` connects to any MCP-compatible client. The same Bearer-token key works for the REST API, the MCP server, and the CLI agent — one auth flow, three surfaces. No demo, no sales call, no separate integration project.