preqo has its own MCP server. Connect Claude, ChatGPT or any MCP client to set up projects, import price lists as units, upload exposés and query leads, with exactly the rights you have in the app.
Example chat: a user asks the assistant to import a price list as units for House B and to send the exposé to every lead interested in 3 rooms. The assistant calls the preqo tools import_units, link_expose and send_message, and House B appears with 16 units on 4 floors.
A lot of sales work is copying data from one place to another.
Units, sizes and prices are copied by hand from a spreadsheet.
Which leads in House B are ready for a consultation? Someone has to build a filter first.
Every new client of an agency means the same clicks for funnels, sequences and users.
Every operation of preqo is also a tool of the MCP server.
Units of house A in Quartier Lichtgarten: 23 units on 5 floors, each marked as available, reserved or sold. While you watch, units get reserved and sold.
The MCP server uses the same operations, rights and logs as the app.
One endpoint for every client, nothing to install locally.
Clients register themselves and sign in with your preqo account. No keys pasted into a chat.
The server checks the same rights and workspace boundaries as the app.
Every change through MCP is logged with user, client and time.
Leads, projects, units, funnels, exposés and more. One registry powers app, REST API and MCP.
Prefer code? The same operations are available as a REST API with an OpenAPI description.
Add the preqo MCP address as a connector in your client.
Log in with your preqo account and confirm the access.
Describe the task in plain language.
Review the result in the app, every change is in the audit log.
Any client that supports MCP over Streamable HTTP with OAuth can connect.
preqo is built in Germany for the rules of the EU. Our data processing agreements and the list of sub-processors are public.
Database, logins and files run on Supabase in the AWS region eu-central-1. The app runs in the Frankfurt region of Vercel.
AI calls, recordings and messages need documented consent. Every consent stores the text version and the time.
The AI assistant says at the start of every call that it is an AI, in line with the EU AI Act.
Roles and row level security separate workspaces. Changes are logged, and deletions are final and documented.
The Model Context Protocol is an open standard that lets AI assistants such as Claude use tools. The preqo MCP server offers the operations of the platform as such tools.
Claude Desktop, claude.ai, Claude Code, Cursor and every other client that supports MCP over Streamable HTTP with OAuth, including ChatGPT where connectors are available.
No. The MCP server works with your login and your role. Rights and workspace boundaries are checked on every call, in the application and in the database.
No. Clients sign in through OAuth 2.1 with your preqo account. For server to server integrations the REST API is available with API keys.
Yes. Every change is written to the audit log with the user, the MCP client and the time.
Yes. Agencies set up client workspaces, import units and create funnels from one chat. Agency access is logged like in the app.
Create a free account and add the MCP server in a minute.