The whole idea, in plain language.

An ontology is a structured model of how a business like yours works — so your AI starts out understanding your world instead of a blank page. Think of it as the difference between a stranger and a colleague. A stranger you brief from scratch every time. A colleague already knows that businesses like yours have clients, that clients have projects, that projects have deadlines and documents, and that a proposal is a kind of document with an amount and a due date.

Because it understands those shapes, it can connect things the way you do in your head, instead of treating everything as one big undifferentiated pile of text. Inside the ontology, that know-how is held as three kinds of thing: blueprints (the templates for the documents you produce), skills (the steps for getting a piece of work done), and advisors (the expertise and judgment that make the output good).

One important thing: on its own, an ontology is a model of a business like yours — a generic map of how a business in your line of work is put together. It isn’t a picture of your business yet. It’s the knowledge layer that makes the work possible. Making it truly yours comes later.

If the ontology is the knowledge, Chief Staffer is the engine that puts it to work. Two distinct things, and the distinction is worth holding onto because it’s also how the pricing is drawn.

The ontology is knowing about a business: the blueprints, skills, and advisors for work like yours. You can have that on its own.

Chief Staffer is the engine that knows your world. As you work, it builds and keeps a living memory of your actual business — a knowledge graph of your real clients, your real projects, what you’ve promised, what you decided and why, and where each fact came from. The ontology tells it the shapes; the knowledge graph holds your particulars. Every time you work, the memory fills in a little more, and it picks up where you left off across sessions.

Put plainly: the knowledge on its own (the ontology) is the $100 product. The engine that runs the work and keeps the living memory of your business (Chief Staffer) is the $300 product, and it comes with one specialization included. The engine is the thing that actually produces grounded work, with sources on demand and gaps named rather than invented.

Chief Staffer isn’t another app to learn or another website to log into. It runs inside the AI app you already use, connected through an open standard called MCP. If you can have a conversation with your AI, you already know how to use it. You ask the way you’d ask a capable colleague, and finished, grounded work comes back.

It works with the common AI apps people already have: Claude, ChatGPT, Cursor, VS Code, Windsurf, and Goose. And if you’d rather run a model on your own hardware, it works with local models too — the quality of the work tracks the capability of whichever model you give it.

Two things follow from this, and they matter. First, it’s local-first and private: your business memory lives on your Mac, not in some company’s cloud. Your data stays with you. Second, you bring your own model tokens, which is just a plain way of saying you use your own AI subscription or account to power the thinking. Nothing about your business gets routed through us, and there’s no surprise bill stacked on top of the price.

You never operate Chief Staffer directly. You work with your own AI the way you always do, and Chief Staffer works behind it. In practice there are just two moves.

Tell it what to remember. When something matters, you tell your AI to save it with Chief Staffer — and it does, drawing from whatever it can already reach: your conversation, a document, an email, a spreadsheet, your calendar, your notes, your other tools. One move, any source — there’s no separate procedure for each. Nothing is saved automatically, so you can talk freely; only what you ask to keep is kept. Forgot to capture something from an earlier chat? Just ask it to add that conversation now. And if a fact is ever wrong, you fix it in a sentence — or point it at the file that sets it straight — and it updates, keeping the old version in the record.

Ask for the work. You think in terms of what you want, not how it gets made. Say “draft the Hartman renewal with Chief,” and it pulls the right blueprint, follows the right steps, and brings in the right advisor — on its own, grounded in your facts. The method is already built in. You bring the goal.

Every business has shape in common, but the work differs. A software team, a consultancy, and a content studio all have clients and projects, but a sprint isn’t a retainer and a retainer isn’t an editorial calendar. A specialization is how you tell Chief Staffer which kind of work you do, so it tunes the whole business model to match.

You pick the specialization that matches the work in front of you, for example Software Development, Consulting, or Content Production. It reshapes the documents, the vocabulary, and the advisors around your actual practice. There’s a General business default to start from, and you can add more than one if your work spans a few areas. Chief Staffer comes with one specialization included; more are $50 each.

A specialization gets you most of the way, but your practice has its own shape that no general model quite captures: your own document types, your own way of pricing, your own language for the work. That’s what authoring is for.

Authoring is a premium add-on for Chief Staffer. Instead of fitting your business to a generic model, Chief Staffer reads your own documents and authors a custom one — your own blueprints, your own advisors, your own language and the way your work actually connects — so the model becomes a true picture of how your business runs. The point is to give your AI a standing brief: it already knows how you operate, so it doesn’t have to work it out from scratch every time you ask. And it keeps learning as you work, so the fit gets closer over time rather than going stale.

Most people start without it — the specializations cover the common cases well. Authoring is there for when the generic isn’t enough, at $200 on top of Chief Staffer.