Run your business from Claude — accounting included (Malaysia)

The stack: your agent in the middle, your books underneath

Documents arrive where they always did — email, WhatsApp, a shared Drive, a photo at the counter. In the middle sits an agent: Claude Code, claude.ai, Claude Desktop, ChatGPT, Cursor. Underneath sit the systems that have to end up correct, and the ledger is the least forgiving.

Your agent reads a supplier bill out of the inbox and needs somewhere to put it. That somewhere is the accounting box.

Taokeh is built to be that box. It is a Malaysian SME ledger — sales, purchases, bank, stock, payroll, SST and e-invoice on one set of books — with an official Model Context Protocol server at taokeh.my/mcp, 64 tools, live since 16 July 2026. The inputs are not ours: email, WhatsApp and Drive reach your agent through your agent’s own connectors, never through Taokeh.

So the only layer you buy from us is the bottom one. Your agent, your model and your input connectors stay yours to pick and to change.

What does “an accounting backend for an AI” actually mean?

The ledger publishes a tool surface your agent calls directly — not a website a human clicks through, not a nightly file export. Three properties make it real.

• Live books. Ask for the aging or the cash position and the answer comes out of the ledger as it stands — not a copy, not last night’s snapshot.
• Work goes back the other way. The agent files an expense off a receipt, a bill off a supplier PDF, an invoice off an instruction.
• The server enforces the boundary, not the prompt. What the agent may do is fixed by the tools that exist, not by how it was asked.

A chatbot in an accounting app is not the same thing

It knows one app, it lives in that app’s chat window, and you cannot point it at your inbox. A backend is what your own agent reaches, alongside everything else it already reaches.

What can the AI actually do against Taokeh — 39 reads and 25 writes?

The server exposes 64 tools: 39 that read and 25 that write. The writes are narrower on purpose.

• The reads cover what you would ask a bookkeeper: the day’s and month’s trading, cash across every account, an eight-week cash forecast, profit and loss, balance sheet, SST position and e-invoice status, receivables and payables aging, sales by channel, stock on hand and the reorder list, document search, and the bank rows still waiting for review.
• Six writes file drafts — expense, bill, invoice, quote, customer receipt, and a contact read off a name card.
• One revises a pending draft the agent already filed, so a misread can be corrected before you look at it.
• One imports a bank statement into the review queue.
• One proposes categories for bank rows, as suggestions the owner sees.
• One issues an upload link for a file too large to inline.
• An owner holding two or more companies under one sign-in can also connect a separate, owner-only group connector: 3 read-only tools for group figures, counted apart from these 64. It never appears on a normal connection.

The safety contract is structural, not a setting

That list is the whole write surface, and none of it posts — a revision changes a pending draft, never the ledger. Reads answer immediately, because a read changes nothing. Writes wait for a human to approve them.

The missing capabilities are missing by construction — they cannot be misused, whatever an agent is told to do. There is no tool at all that:

• posts to the ledger,
• amends or voids a posted document,
• deletes anything,
• confirms a bank row,
• or sends an email.

Your documents stay on your own model

On MCP paths the Taokeh server never calls a language model. Your agent reads the receipt, the statement or the name card and sends structured fields; our server validates them, re-computes every line and total itself, and files a draft.

The reasoning runs on your subscription with your choice of model — your documents are not fanned out to some third model we picked for you.

How do I connect it to my agent?

Two routes, depending on where your agent lives.

• claude.ai and ChatGPT: add https://taokeh.my/mcp as a custom connector. The client registers itself, you sign in at Taokeh, and a consent screen names the company and the scope before anything is granted — no key to copy, no config file to edit.
• Claude Code, Claude Desktop, Cursor and other command-line clients: generate a personal access token inside Taokeh (they look like tkm_…) and send it as an Authorization bearer header. Tokens are stored only as a hash, scoped to one company, and revocable at any time.
• Both routes need the connector add-on switched on for the company. Only admin and bookkeeper seats can connect or use it.
• The transport is Streamable HTTP; auth is OAuth 2.1 with dynamic client registration and PKCE, or the bearer token.

Choosing the scope, and cutting access off

On the OAuth route you choose the scope on the consent screen: read-only, or read plus drafts. You can reconnect later at a different scope, and disconnecting an assistant stops its access immediately.

The step-by-step version, with screenshots of where each setting lives, is the sibling guide at the foot of this page; the endpoint reference lives on the MCP server page.

What stays human in this arrangement?

Three things, deliberately.

• Approvals. Every draft waits for a person to check it against the original document and tap approve — from the review strip inside Taokeh, or from a one-tap approval link that still requires you signed in as admin or bookkeeper.
• Money. Nothing the agent does moves funds, settles an invoice or confirms a bank line. A payment is recorded when a human records it.
• Anything outward. The connector emails nobody and contacts nobody on your behalf. Ask for an invoice PDF and it hands you a short-lived signed link, not a sent message.

Why the read/write asymmetry is worth keeping

Your agent may draft the chase email from the figures it read — but pressing send is your move, in your own mail client, not through us. An agent that reads everything and writes only drafts is safe to give a lot of latitude; an agent that could post would need watching constantly, which defeats the point of having it.

Why does jurisdiction depth matter more than the tool count?

The hard part of Malaysian books is not reading them — it is being right about Malaysian rules. A ledger built for another market can expose a perfectly good connector and still leave you doing the local work by hand:

• LHDN e-Invoice submission and consolidation through MyInvois.
• SST registration thresholds and the SST-02 return.
• Payroll that computes PCB, EPF, SOCSO and EIS to the sen against the current schedules.
• EA forms at year end.
• An interface your staff can actually use in Bahasa Malaysia.

The test when you compare options

An agent cannot invent those — they are compliance, not reasoning. Taokeh carries them in the ledger itself: ask for your SST position or your e-invoice status and you get a real answer, not a number assembled from a spreadsheet you maintain.

So ask whether the statutory work happens inside the box or beside it. If the local filing, the payroll schedules and the e-invoice submission sit outside the system your agent can reach, your agent is only helping with the easy half.

What does this cost?

Confirm the current figures on the pricing page before you rely on them.

• RM29 a month as at August 2026, available on any Taokeh plan rather than reserved for a top tier.
• Included in the 14-day free trial, so you can connect an agent and see how a draft feels before paying anything.
• Flat, with no per-question metering. You bring your own AI subscription, so the reasoning costs whatever you already pay Anthropic or OpenAI — no usage bill from us grows with how much you ask.
• Taokeh’s in-app document scan is a separate, metered feature — not the same thing as asking a question through the connector.
• The add-on price sits on top of the base subscription, which is priced separately.

Frequently asked questions

Updated August 2026

This guide is general information for Malaysian SMEs, not tax, legal or accounting advice. Always confirm current rules and figures with the relevant authority or your own adviser.

Related guides

• Connect Claude or ChatGPT to your accounting books (Malaysia, 2026)
• AI accounting software in Malaysia: what it actually means
• The Ask Taokeh connector — how it works for your business
• The Taokeh MCP server (developer documentation)

Related sources

• Model Context Protocol (open standard)
• The Taokeh MCP server — endpoint, auth and the full tool list