Your org chart can now be read by AI

Drop an AI agent into most companies and watch it get lost. The knowledge it needs to be useful — who does what, which team owns which outcome, where decisions are made — is scattered across a wiki, a few drives, an org chart buried in last quarter’s slides, and a fair amount of nobody-wrote-it-down. So the agent guesses. It pattern-matches from titles, infers a hierarchy that may not exist, and hands you a confident answer built on sand.

That’s not the agent’s fault. It arrived somewhere with no map.

A little while ago we made your organization conversational — you can now talk to your Peerdom organization in plain language, with whichever AI you trust, and get a straight answer in seconds. That solved one half of the problem: getting answers out of your own organization, for you.

This is the other half. Not you asking your organization questions — but any agent, anywhere, being able to read your organization and reason about it correctly, without you wiring up a thing.

A shared language for organizational knowledge

For an agent to feel at home in an organization, it needs the structure handed to it in a shape it already understands. Not scraped, not guessed — stated.

That shape now has a name: the Open Knowledge Format, or OKF. It’s an open standard, published by Google and released under a permissive open-source license, for describing organizational knowledge in a way both people and AI agents can read. The whole specification is deliberately small: a folder of plain Markdown files, one file per “concept,” linked to one another — nothing you need an account, an SDK, or anyone’s permission to produce or consume. If you can read a text file, you can read an OKF bundle.

We could have invented our own format for org structure. We chose not to. A data format is only worth the number of parties who speak it, and a brand-new one nobody else recognizes helps no one. By adopting a standard a broad ecosystem is rallying behind, the agents your people already use arrive pre-fluent in the shape of an organization. We’d rather be an excellent citizen of an open standard than the sole speaker of a private one.

What an agent sees

Here’s the quiet advantage: Peerdom is already a curated, structured model of an organization. The hard part of OKF — turning scattered, half-documented knowledge into clean, linked concepts — is the part you’ve already done every time you mapped a circle or filled a role.

So the translation is natural. Your circles, roles, the people who hold them, your goals, and your pages each become a concept an agent can read and follow:

• It starts at the top and sees the whole organization at a glance.
• It follows a link into a circle, and sees the roles inside it and who fills them.
• It reads a role and understands its purpose, its goals, and how it connects to the rest.

And it reads all of it in your own words — if you call your circles “teams” or “pods,” the bundle says teams and pods. The agent meets your organization on your terms, not a generic template’s.

Make it public, and it’s part of the agentic web

This is the part we’re most excited about.

Any organization that makes its Peerdom map public already shares it openly with the world. From now on, a public map isn’t just a page people can visit — it’s an agent-readable bundle the whole agentic web can understand. No integration, no key, no setup. The moment your map is public, an agent can read it, reason about it, and represent your organization faithfully to whoever’s asking.

Put plainly: Peerdom is where agents read and reason about how your organization actually works.

For organizations that practice transparency on purpose — and many of the ones we work with do — that’s a meaningful thing. The way you’ve chosen to organize becomes legible to the tools the rest of the world is starting to think with.

You stay in control

Legibility is only a good idea with a clear edge around it, so the same controls you already have apply:

• Public is a choice you make. Nothing is exposed until you make your map public — and you can make it private again at any time.
• People are opt-in. A public bundle describes your structure; individual people appear only if you’ve chosen to show them, exactly as with your public map and embeds today.
• Private stays private. If your organization isn’t public, its knowledge stays behind your door — and you can still hand it to a specific assistant, on your terms, through the MCP server and your own API key.

Two ways in, one principle. Conversational access, through MCP, is private and authenticated — talk to your organization. Legible access, through OKF, is open and readable — let any agent understand it. You decide which doors are open.

Why open, every time

We keep making the same choice — an open standard over a walled garden — and it’s worth saying why, because it’s not an accident.

The organizations Peerdom serves tend to believe that clarity and autonomy beat control: that people do their best work when they can see the whole picture and act on it. A closed format, readable only by one vendor’s tools, runs against that grain. An open one hands the choice back — which agent, which model, which infrastructure, on whose terms. We’d rather build for that world and earn our place in it.

Your organization isn’t only something you operate, or even something you can now talk to. It’s becoming something the agentic world can genuinely understand — and that understanding starts from the structure you’ve already built.