Trips
Operations on trips.
Xpense MCP server
The Model Context Protocol (MCP) is an open standard that lets AI assistants and agents work with business systems. Xpense has an MCP server, so an assistant that speaks MCP can work with Xpense, the travel and expense management platform.
How a call works
Sign-in and company binding
An assistant does not get an account of its own. It acts for one signed-in employee, in one company.
The employee signs in
The assistant holds the employee's own sign-in token.
The call carries the token
Every MCP call carries that token.
Bound to one company
The call is bound to one company.
Same permissions
The assistant can do only what that employee can do.
What the tools cover
The tools cover Xpense operations such as these.
Expenses
Operations on expenses.
Advances
Operations on advances.
Reports
Operations on reports.
Who it is for
The MCP server is for companies that already use an AI assistant that speaks MCP, or that build their own agents, and want those tools to work with Xpense.
Because every call carries the employee's own sign-in token and is bound to one company, the access rules you already have in Xpense keep applying. Access in Xpense is role-based.
Book a demo-
The employees list in the portal, with demo data.
Questions about MCP
What is the Xpense MCP server?
The Model Context Protocol (MCP) is an open standard for connecting AI assistants and agents to business systems. Xpense has an MCP server, so assistants and agents that speak the protocol can work with Xpense.
How is access controlled over MCP?
Every MCP call carries the employee's own sign-in token and is bound to one company. An assistant can do only what that employee can do.
What can an assistant do through MCP?
The tools cover Xpense operations such as trips, expenses, advances and reports.
Who is the MCP server for?
Teams that already use an assistant that speaks MCP, or are building their own agents, and want it to work with Xpense under each employee's own access.