Delegate the work.Keep the seal.
Chancelry installs a private AI operator on your own machine, fits it to how your business runs, and stays responsible for it afterwards.
Request a consultationAn operator you run from your phone
The system is built on Claude Code. A lead orchestrator reads your messages on Telegram, passes each piece of work to the specialist agent suited to it, and reports back. Several specialists can work at once.
It can run commands, edit files, push commits and send messages. It keeps a written memory of decisions and context, searchable across sessions, so a new session picks up what the last one settled.
It isn’t hosted anywhere. It runs on your machine, under your own Claude plan.
We run the same design on our own work.
The name comes from the chancellery, the office that held a ruler’s seal and acted with their authority.
What you have at the end
- An orchestrator installed on your machine and running on your own Claude plan.
- Specialist agents set up for the work your business does, with the routing written down where you can read it.
- A Telegram channel that only the accounts you approve can reach.
- A memory in plain files, started with how you work.
- A safety gate in place before the operator is given real work.
- A technical reference and a handoff note for each project it works on.
- Support afterwards, from the people who configured it.
How an engagement runs
Consultation. We learn how your business runs and whether this suits it. Either side can stop here.
Scope. We agree what the operator will handle, which machine it lives on, which accounts it can touch, and what it must never do.
Installation. On your machine. The safety gate goes in before the operator is given anything real.
Configuration. Specialists, routing, memory and rules, written for your business.
Handover. You run it from Telegram while we watch it work.
Support. We stay responsible for it. When something breaks, it gets fixed, and we add a check for that fault.
What’s on your machine at handover
At handover, the way the operator works is written down in files on your machine.
- The rules it follows, in a plain-text file you can change.
- Its memory, as plain files you can read and correct.
- A technical reference for each project it works on, and a handoff note recording where the work stopped and what comes next.
- An audit log of what it did.
None of it needs us to open it. It can also be backed up to a private GitHub repository if you choose. The Claude plan it runs on is yours, paid to Anthropic directly.
It checks before it acts
Every shell command and every file edit passes through a safety gate before it runs.
Some commands are refused outright. A force push to your main branch is one of them, and no approval overrides it.
Others wait for you. A recursive delete or a dropped database table stops until you approve that exact command. An approval covers one use and lapses after 30 days.
Everything else runs, and each action is recorded in an audit log.
Its rules sit in a plain-text file you can read and change. So does its memory. Only the Telegram accounts you approve can reach it.
What the gate is not
The gate is a check inside Claude Code rather than operating-system enforcement. It hasn’t had an independent security audit.
Put a command to the gate
An illustration of the gate’s rules. It isn’t connected to a live system.
git push --force origin main
Refused. No approval overrides it.
Who it’s for
- Owners and operators who already delegate, and want more of the work handled from their phone, inside limits they set.
- People comfortable in a terminal, who want to see how the thing they rely on works.
- Anyone with a Mac that can stay on, a paid Claude plan and a Telegram account.
Who it isn’t for
- Beginners, or anyone looking for an app to download. There’s no one-click install and no graphical setup.
- Anyone who needs it answering while their own machine is switched off.
- Regulated environments that require an independently audited stack.
Questions
Is this software I subscribe to?
No. It’s a service. We install the system, configure it for your business and support it afterwards. The software runs on your machine, under a Claude plan you pay Anthropic for directly.
What do I need before we start?
A Mac that can stay on, a paid Claude plan, a Telegram account, and enough comfort with a terminal to follow what’s being set up.
Where does my information go?
The operator runs on your machine. Conversation history stays there. The work itself is done by Claude, so it travels to Anthropic under your own plan. Your Telegram messages pass through Telegram’s servers, as with any Telegram bot. Memory and project files can also be backed up to a private GitHub repository if you choose.
Could it do something I didn’t intend?
It acts on what you tell it, and it can make mistakes. That’s why the gate exists. Commands that can’t be undone are either refused or held for your approval, and every action is logged. The gate has limits. A process started outside Claude Code isn’t checked, and someone who has already gained the ability to run code on your machine has known ways round it.
Has the safety gate been independently audited?
No. It hasn’t been through a third-party security audit. If your sector requires one, this isn’t the right fit yet.
What happens if my machine is off?
It stops. It runs on your machine, so a Mac that’s shut down or asleep can’t answer, and it won’t bring itself back unattended. We can set up an outside check that messages you when it goes quiet. Expect that alert within hours, not minutes.
Why isn’t there a price here?
Every engagement is scoped to one business. We discuss the fee at the consultation.
Can I change what it does after handover?
Yes. Its rules are a plain-text file you can edit, and its memory is plain files you can read and correct.
Will it act without asking me?
For ordinary work, yes. Commands that can’t be undone are held until you approve that exact command, or refused outright. Everything it does is logged.
What’s left if I stop using it?
The files. Its rules, memory, handoff notes and project references stay on your machine, and in your private GitHub repository if you back them up there. The Claude plan is between you and Anthropic.
Start with a conversation
Tell us how your business runs and what you’d hand over first. If Chancelry suits it, we’ll say so. If it doesn’t, we’ll say that too.
Preview build. This form does not send anything yet.