Frank has nothing to tell you most of the time.
Frank
You didn’t ask — Mudo noticed
The relationship Frank uses here is a verified one: a programme change reaches resourcing, fees and invoicing, and what is claimable this month changes. He names a consequence, never a figure, a forecast or a client.
He is composed by rule, not generated.
Frank cannot state a figure your records do not contain. That is a smaller claim than most software makes about its assistant, and it is the reason the numbers in his briefing can be acted on.
The same records, read by the same rule. His briefing and the screens cannot disagree.
In short
AvailableAvailable now to practices on a qualifying plan.- What it is
- Mudo's briefing voice. Frank composes a short daily briefing from the practice's own records: what is due, what needs attention, and where the commercial position stands. He can read it aloud.
- Who it is for
- Directors and practice managers who want the state of the practice in a sentence before the first meeting, rather than after opening four screens.
- What is different
- Frank's briefing is composed by rule from your records, not generated by a language model. He cannot state a figure the data does not contain, which is why the numbers can be acted on.
- Availability
- Available on every plan, including the free Starter tier.
- How it connects
- Frank reads the same records the Brain draws and the Project Pod holds, so his briefing and the screens never disagree.
Part of Intelligence.
Verified against the Mudo application on .
What a briefing sounds like
Frank speaks in short, plain sentences with real figures in them. The shapes are fixed; the values come from your practice.
He will tell you how many tasks are due today, which project is on the radar and why, what is at the top of the list, and where the commercial position sits. Then he stops. He does not loop, and he does not fill silence.
Why "never invents a number" is a verifiable claim
Most software that says this is describing an intention. In Mudo it is a property of the code: the module that composes the briefing has no imports at all. No model client, no HTTP call, nothing that could return a fabricated figure. It formats values it was handed.
That is a smaller claim than "AI-powered insights", and a much more useful one.
Where generative AI is used, and where it is not
Mudo does use generative models in other parts of the product. The briefing is not one of them, and the distinction is maintained deliberately rather than blurred for marketing.
Common questions
What does Frank do?
Frank reads the practice's projects, programme, time and financial records and composes a short briefing: tasks due today, projects needing attention, the top of the priority list, and the commercial position. He can read it aloud, and the text is always available as well.
Is Frank an AI chatbot?
Not for the briefing. The daily briefing is composed by rule from your own records, which is what makes the figures trustworthy. Mudo does use generative AI elsewhere in the product, and where it does, it says so.
Can Frank get a number wrong?
He cannot invent one. The briefing module composes sentences from values it is given and has no path to a language model, so a figure in a briefing is a figure from your records. If the underlying record is wrong, the briefing will repeat it.
Does Frank have a face?
No. Frank is a presence and a voice, a small indicator and a line of text. There is no avatar, no character illustration and no anthropomorphic animation.
What this page is based on
Every capability described here was checked against the running application before publication. These are the parts of it you can open for yourself.
- /practice/frank-intelligenceFrank's own screen in the application.
Most mornings, silence is the correct briefing.
Free for solo practices: one user, two projects, no card.