Clio.
Shared business memory
Keep the knowledge your business has already earned.
One of fifteen tools the studio built for its own work. We build the same kind of app for clients, shaped around the job they need doing.
- Made for
- Keeping every AI tool working from the same knowledge
- Shape
- Runs quietly in the background, with a Mac app to browse it
- Data
- Stored locally; connected AI and optional external embeddings can send content to providers
What Clio does.
Most AI tools forget the useful context when a session ends. Clio gives the whole studio one shared memory, so a decision made on Monday is still available on Friday, whichever tool we open next.
It runs quietly in the background, capturing the useful parts of each session and linking related knowledge. The records are stored on our own machines. Connected AI tools may send retrieved context to their providers.
Give your team access to the knowledge that usually lives in one person's head: past decisions, important context and the reason an earlier approach was ruled out.
Why we built Clio.
We use AI tools all day, and for a long time each one kept its own memory. A decision made in one session simply didn't exist in the next, or in any other tool. We were re-explaining the same project, the same preferences, the same constraints, over and over.
There was a second worry underneath: when your working memory lives inside someone else's product, it quietly becomes a reason you can't leave. We wanted the studio's accumulated knowledge to be ours: one place, on our own machines, that every tool can use and add to.
The moment that settled it: watching ourselves explain the same project to the same AI for the third time in one week.
How Clio earns its place in a working day.
Work as normal
Start a session with any AI tool we use. The relevant background (decisions, constraints, preferences) is already loaded before a word is typed.
The session ends
The useful parts are distilled and filed automatically, along with a short receipt of what was done. Nobody writes anything up.
Clio organises it
Related records are linked together in the background. Anything uncertain waits for a person to check.
The answer stays available
Months later, ask why a decision was made and find the date, reasoning and options that were ruled out.
Less time spent repeating context
Sessions start with the relevant background instead of a blank page, giving us back the opening minutes of each task.
We can switch tools freely
The memory belongs to us, not to any one AI provider. When a better tool comes out, we move to it and the knowledge comes too.
Handovers got easier
Picking up a colleague's work starts with a brief of the decisions, constraints and recent activity, assembled in seconds, not a meeting.
Made to fit the work it needs to do.
Clio is the foundation several of our other tools stand on, so we built it deliberately boring: a local filing system that can connect to AI services. Optional external embeddings send text to the configured provider; connected AI tools have their own processing and costs. For something everything else plugs into, predictable beats clever.
It's also the clearest example of how we think about ownership. The clever parts (the filing, the distilling, the linking) are replaceable. The knowledge itself is kept in a plain, readable form we could open with or without any particular AI. That separation is the design.
AI files each new record, condenses long working sessions and brings related knowledge together. A person checks anything uncertain before it becomes part of the shared record.
Could your business use something like this?
Tell us the job in plain English. We'll say what would solve it, what it would cost, and whether a desktop app is the right shape for it.