Skip to content

Orbiz MCP · Claude

Connect Claude to your business software, and let it do the work as you

Claude is good at turning a sentence into a finished piece of work, as long as it can reach the system the work lives in. Orbiz MCP is that reach. You add your organisation to Claude once as a custom connector, approve what Claude may do on an Orbiz page, and from then on a request like “log three hours on the Acme redesign” or “who is waiting on my leave approval?” is carried out in Orbiz under your name. This page covers what Claude needs on its side, what it can then do, and where it stops.

Set up

Connecting Claude to Orbiz, step by step

No code and nothing to install. The connection address is inside your Orbiz organisation, not on this page.

  1. In Orbiz, open Settings → Connected AI apps and copy your organisation’s connection address. It is shown only there, with a copy button.
  2. In Claude, open Customize → Connectors, choose Add, then Add custom connector, give it a name and paste the address.
  3. On a Claude Team or Enterprise plan, an Owner adds the connector for the organisation first; each member then finds it under Customize → Connectors and presses Connect.
  4. Claude sends you to Orbiz. Sign in as usual, tick what Claude may read and what it may change, tool by tool, and approve.
  5. In a conversation, switch the connector on from the + menu and ask for the work.

The steps on the Claude side follow its own documentation as read on 4 October 2026. Menus move; the exact steps for today are always on the Settings screen inside Orbiz.

What Claude can do in Orbiz once connected

Orbiz gives a connected assistant 81 actions across 9 of its tools, plus a handful that belong to no single tool: who you are in Orbiz, finding a colleague by name, and searching and fetching content. In practice that means Claude can do the following, when your role allows it:

  • In Cadence: report a bug, create, update, move and close issues, comment, and create, start and complete sprints.
  • In Trackline: create and update projects and milestones, log hours and total them.
  • In AMS: read your day, timesheet and leave balance, request leave, and decide your team’s requests if you approve leave.
  • In HRMS: list your HR tasks, jobs and candidates, complete a task step, add a candidate and move one on.
  • In Ledgr: read your own pay summary and payslips; payroll managers can propose a salary revision.
  • In Notely, Filespace and Grabr: notes and checklists, files and folders, and your recordings with their share links.

Each tool’s page says exactly what an assistant can reach there and what it cannot. The fuller write-ups are the pages on the HR MCP server, the payroll MCP server, the time tracking MCP server and the project management MCP server.

What you need on the Claude side

Custom connectors are part of Claude itself, not something Orbiz installs. Claude’s own help pages list them for the Free, Pro, Max, Team and Enterprise plans, with the Free plan limited to a single custom connector. On Pro and Max you add the connector yourself. On Team and Enterprise an Owner adds it for the organisation once, and after that every member connects individually, which is what keeps each person’s Claude limited to what that person can reach in Orbiz.

Orbiz is added as a custom connector, by address. It is not a listing in Claude’s connector directory, and Claude’s help pages say of every custom connector that it is a service Anthropic has not verified. The check that matters here is your own: the address comes from inside your Orbiz organisation, and the approval page is Orbiz’s.

Claude Code: the same connection, from the terminal

Claude Code is the assistant developers run beside their repository, and it connects to Orbiz too. The difference is where it runs: on your own machine, signing in through your browser. Orbiz gives you a ready-made setup line for it on the same Settings screen as the address; you run that line once and approve in the browser window that opens.

Because a local app is asking, the Orbiz approval page adds a warning it does not show for the hosted Claude apps: only approve if you started it. Once connected, the natural uses are in Cadence: report the bug you just reproduced, with the error still in the conversation, and close the issue with a comment when the fix ships.

What you approve, and what an admin controls

The approval page names Claude, your organisation and your account, and lists access tool by tool with reading and changing as separate ticks. You can untick anything before you approve, and it only offers what your role could already do. Changing pay is its own tick with its own warning. A sensible first connection is read-only: see how Claude answers questions about your work, then reconnect with changes.

Afterwards, Settings → Connected AI apps lists your connections and every action taken through them, with the outcome and the time. Admins see the same for the whole organisation, can disconnect any connection, and can switch AI assistants off for everyone with one control. The platform’s wider safeguards are on the security page.

Examples

What to ask Claude

Requests people actually type. Each one maps to actions Orbiz has; the label says whether it reads or changes something.

  • “Turn these meeting notes into issues in the WEB project and put them in the next sprint.”

    CadenceChanges something

  • “Log 3 hours on the Acme redesign for yesterday, against the Beta milestone.”

    TracklineChanges something

  • “Show me the leave requests waiting on my approval.”

    AMSReads only

  • “Add the candidate from this email to the Backend Engineer job, source LinkedIn.”

    HRMSChanges something

  • “What was my net pay in August?”

    LedgrReads only

  • “Save this summary as meeting-notes.md in the Client A folder and share it with Sara.”

    FilespaceChanges something

Staying in control

What keeps Claude within bounds

Every item here is something you can see for yourself on the approval page or in Settings.

You choose what it may do

Connecting sends you to an Orbiz consent page, signed in as yourself. Access is listed tool by tool, reading separate from changing, and you can untick anything before you approve.

It acts as you, in one organisation

A connection belongs to one member in one organisation. The consent page names both, and says the assistant can only see and do what you can on the web, in the areas you allow.

Pay changes are a separate tick

Changing salaries is its own permission with its own warning on the consent page: only tick it if you run payroll.

Every action is on a list

An activity log shows each action with the assistant, the outcome — done, not allowed or failed — and the time. You see your own; admins see the organisation’s and can filter by member.

Disconnect whenever you like

Any member can disconnect an assistant from Settings → Connected AI apps, and it loses access at once. Admins can disconnect anyone’s.

One switch for the whole organisation

An admin can turn AI assistants off for everyone. That stops every assistant and blocks new connections; existing ones work again when the switch goes back on.

Some things have no action at all

There is no action for the Vault, none for clocking in or out, and none that deletes anything. An assistant cannot be talked into what does not exist.

Before you connect

What it does not do

The limits, stated plainly.

  • Claude has no access to the Vault: there is no action for it.
  • Nothing can be deleted through Claude, and it cannot clock anyone in or out.
  • On a Claude Team or Enterprise plan, an Owner has to add the connector before members can connect.
  • Claude’s Free plan allows one custom connector.
  • Orbiz is added by address as a custom connector; it is not listed in Claude’s connector directory.
  • Claude can be wrong. A request that changes something is worth reading back before you rely on it.

Orbiz MCP · Claude

Questions about Claude and Orbiz

Which Claude plans can connect to Orbiz?

Claude’s help pages list custom connectors for the Free, Pro, Max, Team and Enterprise plans; the Free plan allows one custom connector. On Team and Enterprise an Owner adds the connector for the organisation first, then each member connects with their own Orbiz sign-in.

Is Orbiz in Claude’s connector directory?

No. Orbiz is added as a custom connector using your organisation’s own connection address, which you copy from Settings, under Connected AI apps, inside Orbiz. The address is not published anywhere outside the product.

Does everyone in my company share one Claude connection to Orbiz?

No. Each member connects individually and approves access with their own Orbiz account, so Claude acts as that person and reaches only what that person’s role allows. One member’s connection gives another member nothing.

Can I use Claude on my phone with Orbiz?

Yes. A custom connector added to your Claude account is available in Claude on the web, the desktop app and the mobile apps. You switch it on for a conversation from the connectors menu.

See Claude working in your own organisation

We set up a working organisation for you to try, then connect Claude to it with you on a call.