Companies and projects
Each one its own area: brand, customers, finances, decisions, the records behind them.
Salana
Product · Salana Memory
A Salana Memory is plain files: every insight, decision, number, process and piece of knowledge you have, in one folder you own. The Salana App organises that folder specifically for you and your needs, and puts it to work. Delete the app and the memory still works.
The AI tools will keep changing. Claude today, ChatGPT tomorrow, next month something new. What does not change is what you know: how your companies run, what you decided and why, what worked and what did not, who your customers are, what you are working on right now.
That is the asset. The folder holds it as markdown, readable by any model and by you. Not in a vendor's memory, not in a chat history you cannot export. On your disk, in git, yours.
Any model opens the folder and understands you completely, because the context lives in the files, not in the model. Switch models, switch vendors, switch laptops, even switch the app you look at it through. Your Salana Memory comes with you, without an export or a migration.
The value is not in which AI you use this month. It is in the processes, the automations and the tasks that run your company and your life from the folder, and in never having to explain the context more than once.
Everything you do flows into one memory. The agents read it, and the work comes out the other side.
Every chat starts with full context: your brand, your numbers, your history, your rules. You explain something once, it is written down, and every agent after that reads it before it starts. A new chat in a new tool picks up where the last one left off.
The Salana App sets the folder up from an interview with you, then keeps it tidy as it grows. Every agent reads the same conventions first: one index per area, links between notes, and a schema for anything recorded twice.
Each one its own area: brand, customers, finances, decisions, the records behind them.
Daily snapshots of your numbers, imported from your tools, so a question about last month has something to read.
Skills: how you do things, written once, run by you or by an automation.
What runs when, declared as notes, executed by the Salana engine on your machine.
A log of everything done, written by the agents as they work.
Every conversation kept, a daily activity log, and an events feed.
The processes, automations and tasks run from the folder while you work on what moves you forward. Every morning an agent reads your numbers from the memory, drafts what needs drafting, files what needs filing, and tells you what to do today.
Your files, every model, yours forever.