Ask your business a question. Get an answer from your own data.
The SOS + Dist Mgr Assistant is built on Claude AI and wired straight into your system: ask in plain English, and it answers from your live data and the built-in help. Read-only, on your own AI account, with a budget cap you set.
The questions you already ask someone every day.
Data questions and how-do-I questions both work — it reads the same help docs your crew does, so it can explain the system, not just query it.
Designed in, not bolted on.
A lot of vendors are retrofitting a chatbot onto a twenty-year-old system and hoping it guesses right. SOS and Distribution Mgr are newly written — the database schema and the built-in documentation were designed with the Assistant in mind, so it knows exactly where everything lives.
You ask in plain English
“Show me all open POs past their promised date.” “Which jobs are over estimate this month?” Type it the way you’d say it — or attach a screenshot, PDF, or CSV and ask about that.
It reads, never writes
The Assistant looks at your live database and the built-in help docs to work out the answer. Its access is read-only by design — there is deliberately no way for it to change, add, or delete anything.
You get the answer, with the numbers
It answers in plain language with the actual figures from your system, streamed live. Conversations are saved per user, so you can go back to an answer next week.
Five minutes to set up. A dollar cap to sleep on.
Paste your Anthropic API key, pick a monthly budget, click Test Connection — done. The setup screen shows estimated spend against your cap all month, and permissions decide who can use it. Turn it off any time; SOS runs exactly the same without it.
What keeps it trustworthy.
Read-only, by construction
This isn’t a policy or a promise — the Assistant has no write capability at all. It can look things up and explain them; changing your data is impossible for it.
Your own AI account
It runs on your own Anthropic API key. Usage is billed by Anthropic directly to you at their published rates — we never see it and never mark it up.
A budget you set in the app
Set a monthly dollar cap right on the setup screen and watch estimated spend against it. When the cap is reached, the Assistant pauses. A typical question costs a fraction of a cent.
Nothing trains on your data
Your questions and your data don’t train the model, and one user’s conversations never carry into another’s.
You decide who can use it
Access is granted per user under Users & Security — the same role-based permissions as everything else in SOS.
Honest about being AI
The Assistant cites what it looked at, and the app says it plainly: AI can make mistakes, so double-check cited sources. For anything that matters, the numbers are one click away in the real screens.
Asked and answered.
Can the Assistant change my data?
No. It is read-only by construction — it has no ability to add, change, or delete anything in SOS or Distribution Mgr. It can only look up and explain.
What does the Assistant cost?
The Assistant runs on your own Anthropic API key, so usage is billed by Anthropic directly to you at their published rates — typically a fraction of a cent per text question. You set a monthly budget cap inside the app, and the Assistant pauses if it is reached.
Does it learn from our conversations?
No. Nothing you ask trains the model, and conversations do not carry between users. Each user has their own private history, which they or an admin can delete.
Is this bolted onto an old system?
No — and that is the point. SOS and Distribution Mgr are newly written, with a database schema the Assistant was designed alongside. It knows exactly where everything lives, which is why its answers come from your actual records instead of guesswork.
Who can use it?
Whoever you allow. Access is controlled per user under Users & Security, and an admin can enable or disable the Assistant for the whole company at any time.
Ask it something about your own shop.
Book a demo and bring the question you always wish the system could just answer.