Provider REST API

20 actions

Connect Claude to Brex

Read expenses, card and cash transactions, cards, users, departments, budgets and spend limits. Toolspoke puts 20 of its actions behind one MCP endpoint that Claude, Cursor, and Codex all speak.

Connect View pricing

Connection: Provider REST API
Authentication: User token
Actions exposed: 20
What a call bills: Nothing
Adapter: Maintained by Toolspoke

Connected in three steps

1
Install Brex
Open the marketplace in your workspace, add Brex to the project your agents work in, and it appears on the gateway immediately.
2
Connect the credential
Authenticate with user token. Where to get one, and what it has to be able to reach, is the next section.
3
Point your agent at the gateway
Give your client one address, https://toolspoke.com/mcp. Claude Code takes it as a command, Claude and Claude Desktop add it as a custom connector, and Cursor, Codex and VS Code each read it from a config file of their own.

.mcp.json

{
  "mcpServers": {
    "toolspoke": {
      "type": "http",
      "url": "https://toolspoke.com/mcp"
    }
  }
}

One block covers every tool you have installed. Brex shows up in the client as soon as your policy allows it, and so does everything else you install later.

Where the address goes, per client

Claude Code: Run it in your project, then /mcp to sign in
claude mcp add --transport http toolspoke https://toolspoke.com/mcp
Claude and Claude Desktop: Settings, then Connectors, then Add custom connector
https://toolspoke.com/mcp
Cursor: ~/.cursor/mcp.json, or .cursor/mcp.json for one project
{ "mcpServers": { "toolspoke": { "url": "https://toolspoke.com/mcp" } } }
Codex: ~/.codex/config.toml
[mcp_servers.toolspoke] url = "https://toolspoke.com/mcp"
VS Code: .vscode/mcp.json, or the MCP: Add Server command
{ "servers": { "toolspoke": { "type": "http", "url": "https://toolspoke.com/mcp" } } }

What Brex asks for

User token. You provide it once, when you install the connector. Toolspoke encrypts it at rest and decrypts it only for the length of a single call, and the gateway attaches it to the outbound request itself, so it is never part of the arguments an agent sends.

User tokenRequired: Brex → Settings → Developer → Create token (an account admin or card admin must have accepted the Developer API agreement first). Grant the read-only scopes for Expenses, Transactions, Team and Budgets; add the Cards write scope only if lock_card, unlock_card or terminate_card should work, and Expenses write for update_expense_memo.; Paste the token

What Claude can do in Brex

20 actions, each one declared and named by the connector rather than discovered at runtime. A workspace policy grants a person all of them, a hand-picked selection, everything on the read side, everything on the write side, or none.

Reads: 16Reads
Writes: 3Writes
Destructive: 1Destructive

Reads

Fetches data and changes nothing.

get_company
Retrieve the Brex company - legal name, mailing address and account type.
get_me
Retrieve the Brex user this token belongs to.
list_users
List the company's Brex users with their department, location, manager and status. name does a fuzzy match on first name, last name and email.
get_user
Retrieve one Brex user by id.
list_departments
List the company's departments, optionally filtered by name.
list_locations
List the company's locations, optionally filtered by name.
list_expenses
List expenses of every type - card, bill pay and reimbursement - with their status, amount, merchant, memo and receipt compliance. Narrow by purchase date and status first; Brex caps a page at 100. Pass expand (e.g. merchant, user, department) to inline related records instead of ids.
get_expense
Retrieve one expense by id, of any type.
list_card_transactions
List settled card transactions across all card accounts, newest first. Use posted_at_start to bound the window.
list_card_accounts
List the company's card accounts with current balance, available balance, limit and the current statement period.
list_cash_accounts
List the company's Brex business (cash) accounts with their balances. Account and routing numbers are masked in the audit log.
list_cash_transactions
List the transactions on one Brex business (cash) account.
list_cards
List cards, optionally only one user's, with status, limit type, spend controls and last four digits. The card number itself is never returned here.
get_card
Retrieve one card by id, with its owner, status and spend controls.
list_budgets
List the company's budgets with their owners, periods and amounts.
list_spend_limits
List spend limits, optionally only those a given user is a member of, with their amounts, periods and policies.

Writes

Creates or updates something on the other side.

update_expense_memo
Set the memo on a card expense. A write rather than destructive: it changes the expense's description only, and the memo can be set again. Admins and bookkeepers can update any expense; other users only their own.
lock_card
Lock a card so it declines new purchases. A write rather than destructive: unlock_card reverses it. Brex requires a reason.
unlock_card
Unlock a card that lock_card locked, so it can be used again.

Destructive

Deletes or permanently alters something. Worth granting on purpose.

terminate_card
Terminate a card permanently. Destructive because it cannot be undone: a terminated card can never be unlocked, and its holder needs a new card. Use lock_card for anything temporary.

What it will not do

Enforced by the gateway rather than left to convention, which is why each of these can be stated flatly.

It cannot call anything else: The 20 actions above are the whole of it. A call to any other name is refused before it reaches Brex rather than forwarded on, and connecting your account does not add to the list: it is fixed by the connector, not discovered at run time.
It reaches no further than your credential: Toolspoke holds no access to Brex of its own. Every call carries the credential you stored and nothing besides, so whatever that credential cannot reach, this connector cannot reach either.
It never hears from Brex: Nothing is pushed to it. There is no webhook, no subscription and no polling, so this connector cannot notice by itself that something changed in Brex. An agent has to ask.
It does not smooth over provider limits: Toolspoke does not retry, queue or back off around Brex's own rate limits. A call that Brex refuses comes back to the agent as a failed call.

Before you connect it

What can Claude do in Brex?

20 named actions: 16 that only read, 3 that write and 1 that delete or permanently alter something. They include get_company, get_me and list_users. Nothing outside that list is reachable: the connector declares each operation by name rather than proxying whatever an agent asks for.

What credentials does the Brex connector need?

User token. The connector asks for user token. Values are encrypted at rest and attached to the outbound request by the gateway, so they are never part of the arguments an agent sends and never reach the audit log.

Does the Brex connector work with Cursor and Codex, or only Claude?

Any client that speaks MCP, and every one of them gets the same 20 actions. There is a single address, https://toolspoke.com/mcp. Claude Code adds it with claude mcp add --transport http, Claude and Claude Desktop take it as a custom connector in settings, Cursor reads it from .cursor/mcp.json, Codex from ~/.codex/config.toml, and VS Code from .vscode/mcp.json. Each of them signs in to the gateway itself, so there is no key to paste.

What does the Brex connector not do?

The 20 actions above are the whole of it. A call to any other name is refused before it reaches Brex rather than forwarded on, and connecting your account does not add to the list: it is fixed by the connector, not discovered at run time. Toolspoke holds no access to Brex of its own. Every call carries the credential you stored and nothing besides, so whatever that credential cannot reach, this connector cannot reach either. Nothing is pushed to it. There is no webhook, no subscription and no polling, so this connector cannot notice by itself that something changed in Brex. An agent has to ask. Toolspoke does not retry, queue or back off around Brex's own rate limits. A call that Brex refuses comes back to the agent as a failed call.

Can I limit which actions an agent can call?

Yes, in two places. The project switches Brex's actions on and off one at a time, for everyone in the project at once, and the screen groups them by read, write and destructive so turning off everything that deletes is one click. An individual agent key can then be narrowed further, to particular toolkits in a project and to particular actions in a toolkit. Whatever it was granted, a key never reaches a project its owner cannot.

What gets recorded when an agent calls Brex?

Every attempt, with the agent that made it and the person that agent belongs to, the full request payload, the response payload, the status, the duration, and what the call cost in sandbox time and model tokens. Values whose key names a secret are masked out before the record is shown to anyone. An operation the connector marks as not retained never has its response body written at all, so the gateway keeps no second copy of what was read.

More finance connectors

Browse all of them