Ask Cortena in Claude, ChatGPT, Perplexity, Gemini, Grok, or any MCP client.
Connect your favourite LLM to Cortena. Ask about your payables in plain language and get real answers on clean, real-time data. What is open, what needs approval, how your spend is trending.
ClaudeAsk Cortena anything.
app.cortena.ai/api/mcpAn MCP server for your accounts payable.
Connect Cortena to your favourite assistant once, then just ask about your payables in plain language. No exports, no digging through dashboards.
Behind it is your live Cortena data, your invoices, metrics and queues, scoped to what you are allowed to see. Nothing is trained on your data.
app.cortena.ai/api/mcpSee it in action.
Once connected, ask in plain language and get a real answer from your books. Three things teams do every day:
A quick question
Ask in plain language and get a real answer, straight from your live books.
You askWhich invoices are still open?12 invoices are open, €38.940,00 in total. 3 are overdue.
- Nordwerk Bürobedarf€9.800,00
- Lampe Energie AG€7.011,10
- Hetzner Online GmbH€197,42
What needs you
See exactly what is waiting for your review or approval, in one place.
You askAnything that needs me today?Two need review, two are ready for your approval.
Analyse on the fly
Turn a question into analysis: trends, comparisons and a clean table.
You askAnalyse my top 3 vendors this month. How are they evolving?Your top 3 vendors this month, against last month:
An assistant is only as good as its data. Ours is real time, from day one.
Connect an assistant to messy payables and you get confident answers that are quietly wrong. The reasoning is fine, the ground truth is not. So we split the problem: your assistant answers in plain words, Cortena builds the part you cannot leave to chance.
Your assistant understands the question and replies in plain language, over your own data. That part is genuinely good.
Cortena delivers clean, booking-ready data. Right entity, right accounts, right cost centre, right VAT. Same invoice, same result, every time.
Right from the first minute
Invoices are captured and pre-booked as they arrive, so the data is booking-ready from the moment it lands.
Real time
Invoices are pre-booked in seconds. Often the answer is ready before you finish asking.
Same rules, every time
Every invoice coded, every line booked, on your rules. The same invoice gives the same result, not a fresh guess each run.
Answers on truth
Your assistant reasons over correct financial data, not a spreadsheet you hope still holds. That is the difference between guessing and knowing.
The Cortena MCP server, answered.
The Cortena MCP server is a Model Context Protocol endpoint at app.cortena.ai/api/mcp that lets an AI assistant read and act on your accounts payable. You add it as a custom connector in Claude, ChatGPT, Perplexity, Gemini, Grok or any MCP client, then ask about your invoices, metrics and queues in plain language.
You can connect any assistant that supports custom MCP connectors: Claude, ChatGPT, Perplexity, Gemini, Grok and more. You point the client at app.cortena.ai/api/mcp and authorise it over OAuth, scoped to your Cortena workspace.
You can list and search invoices, read your Pulse dashboard metrics, check workflow queues, and get deep links back into the app. The assistant answers from your live Cortena data, filtered to what your account is allowed to see.
Not by default. Connectors are read-only unless you grant a client write access. When you do, writes follow a prepare, confirm and commit flow, so nothing is changed until you approve the exact action.
Yes. Every request is scoped to your tenant and permissions, all activity is logged, and clients connect over OAuth so you can revoke any of them at any time. Data is hosted in the EU and is never used to train AI models.
The MCP server is in beta. It is available to Cortena customers who want to connect their assistant, and we are expanding coverage of tools and clients. Book a demo and we will set up a connector for your workspace.
Talk to your payables.
Book a demo and we will connect your assistant to your Cortena workspace, so you can ask about your invoices, metrics and queues from the assistant you already use.