Manage FreeAgent contacts, invoices, expenses, bank transactions, timeslips, projects and tasks.
FreeAgent ships in the w6w first-party pack. It declares 28 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.freeagentFreeAgent brings the core small-business accounting workflow into a workflow — contacts, invoices, expenses, bank transactions, time tracking (timeslips), projects and tasks. Contact and invoice actions cover the standard create-read-update loop, letting a workflow keep customer records current and generate invoices as a business process requires; invoices always start as Draft, so a dedicated action sends the invoice email once it’s ready. Expense actions record claims against a user, bank transaction actions read transaction history for reconciliation, and project/task/timeslip actions cover logging and reporting billable time.
FreeAgent references a related resource — the contact on an invoice, the task on a timeslip — by its full API URL rather than a bare id, so every action that names one takes a plain id and builds the URL FreeAgent expects internally. Create and update actions accept FreeAgent’s own field names directly for anything beyond the few fields every call needs, so the full breadth of what FreeAgent’s Company API supports on a contact or invoice is reachable without the action surface needing to enumerate every possible field.
This app deliberately leaves out credit notes, estimates, bills, capital assets, payroll, sales tax returns, attachments and the separate Accountancy Practice API (a different product for accountants managing multiple clients), focusing on the core contact, invoice, expense, bank transaction and time-tracking operations most workflows need first.
Three routes to the same 28 actions. The Workflow tab is generated from FreeAgent's own manifest and carries its real ids, so it is copy-pasteable; the Code and CLI examples are the same call for any action on any app, so every app-specific value in them is a blank you fill in.
contact-create Create a new contact (customer or supplier). Provide either an organisation name, or a first and last name.
invoice-create Create an invoice. FreeAgent always creates it as Draft — use Send Invoice Email or a status transition to move it on.
A workflow step names the app and the action, and the editor fills in the
connection when you pick one. This is the Step shape from the
workflow spec, carrying FreeAgent's real ids.
{
"manifestVersion": "2",
"name": "freeagent-example",
"steps": [
{
"id": "bank-transaction-get",
"uses": {
"app": "io.w6w.freeagent",
"action": "bank-transaction-get",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"bankTransactionId": "<bankTransactionId>"
}
}
]
}bank-transaction-get bank-transaction-list contact-create contact-get contact-list +23 more actions available
Every app-specific value here is a blank you have to fill in. An
app action is reached through the connection that authenticates it, so the
address is a connection id, not the app id — and connections belong to your account,
so a public page cannot know yours. Create one for FreeAgent, then fill in
the three blanks: conn_YOUR_CONNECTION_ID, the action key, and the
parameters that action declares. The call itself is real — the shape is transcribed
from the studio's own snippet builder, which prints the same kind of blanks — but
nothing in it is specific to FreeAgent. The Workflow tab is where this app's
real ids are.
npm install @w6w/sdkyarn add @w6w/sdkpnpm add @w6w/sdkdeno add npm:@w6w/sdkimport { W6wClient, isActionRun } from "@w6w/sdk";
// Reads W6W_BASE_URL and W6W_TOKEN from the environment when omitted.
const client = new W6wClient();
const envelope = await client.run({
urn: "conn_YOUR_CONNECTION_ID",
action: "bank-transaction-get",
payload: {
bankTransactionId: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action bank-transaction-get --payload '{"bankTransactionId":"<value>"}' Give an AI agent FreeAgent — without giving it FreeAgent's credentials. One MCP endpoint exposes every app, function and workflow the caller is entitled to, as tools it can discover and run. Access is granted per team while we onboard.
One tool call{
"name": "w6w_invoke",
"arguments": {
"ref": "app:io.w6w.freeagent#bank-transaction-get",
"input": {
"bankTransactionId": "<bankTransactionId>"
}
}
}
Every tool names its target with a single ref. The
app: form above doesn't name a connection at all — the
host resolves which of the caller's FreeAgent connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
The token is attached host-side, at the moment of the call. It is never a tool argument, never in the model's context, and never in a transcript — so a prompt injection has nothing to exfiltrate.
Tools are derived per end user from what that person has actually connected and is entitled to — not one shared bot identity carrying the union of everyone's access.
Multi-step work runs on the workflow engine and returns a run handle the agent can poll — retries, branching and state survive the conversation that started them.
FreeAgent's declared health checks are on the surface too, so an agent can tell "the vendor is down" from "your credential expired" before it burns a retry on either.
The MCP surface is part of the hosted platform. FreeAgent itself is MIT, and the runtime that executes it is source-available (FSL).
FreeAgent declares its own checks, so its health is a property of the app rather than something the host guesses at.
Atlassian Statuspage rollup for status.freeagent.com, with per-component detail. Unauthenticated and unsigned.