What kyte is
kyte is the software a small company runs on: customers, projects and invoices, the processes people follow, the goals they are working toward, and the documents that explain how things are done.
What makes it different is that an agent sits inside it, with access to all of that at once. You can ask for a customer to be created, a process to be started, or a question to be answered from your own documents — in the same window where you would otherwise click.
The five apps
| What it holds | |
|---|---|
| CRM | Organisations, contacts, deals |
| ERP | Customers, projects, invoices, time |
| Processes | The steps your company repeats, and each run through them |
| Goals | Objectives and key results, with the work that belongs to them |
| Knowledge | Documents — written here or in a folder on your disk |
They share one database, so a customer in ERP and an organisation in CRM are the same company, not two records that drift apart.
Your data stays where you put it
kyte reads and writes a database on your own machine. That is not a cache — it is the real thing, and the app works with no connection at all.
When your company works as a team, a server holds the shared copy and each device keeps the part its user is allowed to see. Nothing is stored on someone else's computer unless you chose that during setup.
The agent works as you
Everything the agent does, it does under your name and with your permissions. It cannot see a record you cannot see, and it cannot change one you are not allowed to change.
Anything consequential is asked first. You will see what it intends to do and can say no.
What you can build yourself
The five apps are not a fixed set. You can describe something your company needs — a register of suppliers, a checklist for onboarding, a small tool for a recurring job — and the agent builds it as a working app inside kyte. That mode is called Extend.
Next: What can I use kyte for? — the things people actually do with it. Or go straight to installing it.