Month-end review
Daily spend summaries with top merchants and categories are already in the graph, so finance can ask for the month's pattern and then drill into any day's expenses with brex_expenses.
Who spent what on which Brex card, and against which budget.
Connect Brex with a read-only user token and ask about card spend, individual expenses with their memos and receipts, budgets versus actual spend, and cash balances. Memtro summarises each day's card spend and flags large or out-of-policy expenses.
Ask in plain English from chat, your editor, a scheduled job or your own agent. The model picks the right Brex tool, reads what it needs, and answers with the source cited.
Daily spend summaries with top merchants and categories are already in the graph, so finance can ask for the month's pattern and then drill into any day's expenses with brex_expenses.
An expense flagged out of policy, or any single purchase of 500 or more, becomes a memory naming the cardholder, merchant, budget and memo the hour it appears.
Questions people ask once Brex is connected. In agent mode each one also checks the other systems that know about the same people, companies or identifiers.
Teams that get the most from Brex in Memtro, with the rest of their stack attached.
A record in Brex is rarely the whole story. Connect the systems that hold the rest and the agent cross-references them by default, then remembers what it learned so the next question starts further along.
Using something else for the same job, or both? Each connector has its own tools; connect as many as you like.
Chat in the dashboard, the OpenAI-compatible API, scheduled jobs, and any MCP client: Claude Code, OpenCode, Cursor, Claude Desktop. The Brex tools are the same everywhere.
Set it up as an MCP server →Clients and agents →Seven-day free trial, then one monthly price per organisation. Connectors, jobs and memory are never metered.