Your company brain is a connected folder of files.
We provide the plumbing.

Plain files every team member's AI can read. Sharing, governance, integrations, audit trail, versioning, rollback and export come built in.

Try it free for 7 days, no credit card required

Connecting an assistant — one library, reachable from each team member's Claude, ChatGPT or Cursor
A shared library — notes, prompts and skills, kept as files the company owns

1. Every team member's AI — 2. One library for the company

The problem

Your company already has a brain.
It's just scattered.

Notes, calls, credentials, decisions, the way your best rep works. They live in people's heads and chats, and in one vendor's memory. Each team member has their own slice. Maybe some data is centralized, but access is touch-and-go. And no one can say who changed what.

Connecting it is mostly a sharing and governance problem.

The library

A library, not a black box.

Plain files your team can read, reason about and audit. Curated by people, native to every model, and yours to take to whatever comes next.

Choosing an AI app to connect — Claude, ChatGPT or Cursor

Every team member's AI

Claude, ChatGPT, Cursor: every seat reads and writes the same library over MCP.

A company's library — folders nested inside folders, with a skill and a prompt among the files

One person's method, everyone's

The best rep writes down how they do it. Now every team member's AI does it that way.

A shared library's member list

Walls where you need them

Roles across the library, grants per folder. Only the people on a matter can read it.

Update a doc, and now every agent has the new facts.

The plumbing

Solve for docs and data what is solved for code.

Developers adopted AI fast because everything they touch is versioned. Your library gets the same, without anyone touching git.

Sharing

One library, every team member and agent.

A shared library the whole team reads, writes and runs agent work against.

Governance

Your policy, enforced automatically.

Roles, per-folder grants, separate read and write, and scoped access for each AI connection.

Versioning

See what the AI changed. Undo it.

Every write is a version, flagged when AI made it, and one click from rollback.

Audit

When an answer is wrong, see what it read.

Who changed what, when, and with which tool or model.

Security

One place for an admin to say yes (or no).

Integrations (or "connectors"), credentials and grants live with the library, not in a hundred tools.

Export

An open format no one can retire.

The whole library exports as plain markdown and attachments, ready for whatever comes next.

In the cloud

Dispatch work. Share the results.

Cloud runs complete tasks autonomously using your shared context and credentials. Ad hoc or on a schedule. In secure, throwaway sandboxes. Results come back to your library: refreshed docs, data pulled from your sources, daily or weekly reports.

Many at once, at arm's length.

Cloud runs happen off your machine. Only the result comes home.

Four runs going at once in the cloud, two still working and two finished

Every asset, accounted for.

How each file came to be: the prompts that ran, the models used, what it cost.

Two documents from one process — the step that made each, the model that wrote it, and what it cost

The durable record lives outside of AI, in files the business owns.

Pricing

Solo or team. Your machine or ours.

Jon Hall, FactorIn's founder

Not sure which plan fits? Book a 15‑minute call with the founder.

Book a call

Pro

Your personal library of notes, prompts and skills that works with any AI

$29 / month

 

Start my library

Cloud adds

  • Start autonomous AI work from anywhere
  • Work runs with your laptop closed
  • Run many tasks at once
  • A secure sandbox for every cloud run
  • An audit trail of every step
  • Unlimited notes, prompts and skills
  • Reachable from any AI tool over MCP
  • Can run on your own machine
  • With Cloud: 50 hours of cloud compute per month
  • With Cloud: Cloud workspaces for file- and data-heavy projects
  • With Cloud: Connect your Claude or ChatGPT subscription, or bring your own API keys
  • With Cloud: Or buy credits and run on ours

Team

Your shared library of notes, prompts and skills that works with any AI

$87 / month

 

Start my library

Cloud adds

  • Start autonomous AI work from anywhere
  • Work runs with your laptop closed
  • Run many tasks at once
  • A secure sandbox for every cloud run
  • An audit trail of every step
  • Unlimited notes, prompts and skills — private or shared
  • A personal library for every member, alongside the team's
  • Includes 3 seats
  • Additional seats $29/mo each
  • Reachable from any AI tool over MCP
  • Can run on your own machine
  • Unified team dashboard and controls
  • With Cloud: Additional seats $89/mo each
  • With Cloud: 50 hours of cloud compute per seat (150 hours included)
  • With Cloud: Cloud workspaces for file- and data-heavy projects
  • With Cloud: Connect your Claude or ChatGPT subscription, or bring your own API keys
  • With Cloud: Or buy credits and run on ours

About the same price as Claude or ChatGPT, for a library the company keeps.

Need SSO, custom data terms, dedicated support or volume pricing? Contact us

Frequently asked

Before you ask.

Is this a company brain?
One you can read, yes. So we prefer the term "library." Only neuroscientists understand a brain. A library is systematic, a curated collection of documents and data, versioned and attributed. It's not just where you go for answers, it's also a place you can go to learn.
Why not a shared drive?
A drive can't be reached by every team member's AI, doesn't record which model changed what, and can't roll back an agent's edit. We add that. The files stay plain files.
Do we have to standardize on one AI?
No. Each team member keeps the assistant they already use. They all read the same library over MCP.
Who can see what?
Roles set what someone can do across the library: viewer, editor, member, admin, owner. You can also grant read or write access to specific folders.
What happens to our files if we leave?
Your files are yours. The whole library exports as plain markdown and attachments you can read and write without us. If a better system comes along, you're ready.
Who pays for the AI itself?
You do, on your own subscriptions or API keys — or on credits you buy from us. The plan buys the library and the compute; the model's bill is never bundled into it.

Your team's expertise, in files you own.

Capture it. Share it. Govern it. Take it anywhere.

Start my team library