Vetera MCP is a public server that gives your own AI tool — Claude, ChatGPT, Gemini, Microsoft Copilot, or any MCP-compatible client — read-only access to your Vetera reporting data. Vetera provides the data; your AI tool does the analysis. Connect it once, then ask your AI tool natural-language questions about your clinic's data.
Note: If you don't see Vetera MCP in your Open API access settings, contact your clinic administrator.
What Vetera MCP Does
Vetera MCP gives your AI tool read-only access to your Vetera data. Vetera supplies the data; your AI tool interprets it, summarizes it, and answers follow-up questions.
Vetera MCP covers the following data areas:
Invoicing
Debtors
Payments
Appointments
Clients and patients
Health plans
Clinical activity
Catalog
Vetera MCP works alongside your existing reports in Vetera. It does not replace them.
If you would rather ask questions about your clinic's financial data without connecting an external AI tool, use Ask Vetera instead. Ask Vetera is built into the Reporting section of Vetera and needs no setup. Use Vetera MCP when you want to work in your own AI tool.
Before You Start
Vetera MCP must be enabled for your clinic before you can connect to it.
To use Vetera MCP, you need one of the following:
The Administrator role in Vetera
Membership in your organization's Vetera MCP personnel group, which is created when Vetera MCP is enabled
A shared department with that group
Note: The browser extension does not support Vetera MCP.
Enable Vetera MCP for Your Clinic
Select your clinic location name in the top left corner, then select Settings.
Select General > Integrations > Open API access. This opens the API settings page, where Open API access is one of the sections.
Select Add application.
Find Vetera MCP in the catalog, then select the button beside it. The catalog is in alphabetical order and runs over several pages, so Vetera MCP may not be on the first page. Use the page numbers at the bottom of the dialog.
Select Save.
Vetera MCP now appears in your Applications list, enabled for your organization.
Vetera MCP reaches your data through its own virtual integration user. You do not create or manage this user. To see the user and your endpoint URL, select the list button on the Vetera MCP row.
Connect Your AI Tool
You need your Vetera MCP endpoint URL to connect any AI tool. For clinics in Germany, Austria, and Switzerland, the endpoint is: https://mcp.provetcloud.app/mcp/
For clinics in other regions, use the endpoint URL shown on the Vetera MCP row in your Open API access settings.
If You Have an Individual Claude Pro or Max Plan
Select Settings, then select Connectors under Customize.
Select the plus (+) button, then select Add Custom Connector.
Name the connector Vetera and paste your Vetera MCP endpoint URL.
Select Add.
If Your Organization Uses Claude Team or Enterprise
Only a Claude account owner can add a new custom connector for a Team or Enterprise organization. If you're not the owner, ask them to complete the setup step below first.
Setup (Claude account owner only):
Select Organization Settings > Connectors.
Select Add.
Point to Custom, then select Web.
Paste your Vetera MCP endpoint URL.
Select Add.
Connect (everyone else in the organization):
Select Settings, then select Connectors under Customize.
Find Vetera in the list. It's labeled Custom.
Select Connect.
If You Use ChatGPT
ChatGPT needs Developer mode turned on before you can add Vetera MCP.
Open ChatGPT in a web browser, select your profile, then select Settings > Connectors.
Open Advanced, then turn on Developer mode.
Return to Connectors, then select Create.
Paste your Vetera MCP endpoint URL, then select Create.
If You Have a Google AI Ultra Plan (Gemini Spark)
Gemini custom apps need Gemini Spark, which is part of the Google AI Ultra plan. You can only add a custom app in the Gemini web app, not the mobile app.
Go to gemini.google.com in a web browser.
Select Settings & help at the bottom of the menu on the left, then select Connected apps. If you do not see it, look under Personal Intelligence first.
Under Custom apps for Spark, select Add a custom app.
Paste your Vetera MCP endpoint URL into the MCP server URL field, then select Next.
After you connect in the web app, Vetera also works in the Gemini mobile app.
Note: Custom connected apps are available on personal Google accounts in the United States, in English. Work and school accounts are not supported.
If You Use Microsoft Copilot
Microsoft Copilot has no field for pasting an MCP server URL. Your IT administrator adds Vetera MCP to an agent in Microsoft Copilot Studio instead, and Copilot users reach your Vetera data through that agent.
Note: Building the agent needs a Copilot Studio license. Everyone who uses the agent needs a Microsoft 365 Copilot license.
Your IT administrator does the following:
Sign in to Microsoft Copilot Studio at copilotstudio.microsoft.com.
Open the agent to connect Vetera to, or select Create > New agent.
Go to the agent's Tools page, then select Add a tool.
Select New tool, then select Model Context Protocol.
Enter Vetera as the server name, add a short server description, then paste your Vetera MCP endpoint URL into the Server URL field.
Select OAuth 2.0 as the authentication type. Vetera MCP is added using its server URL alone, with no client ID or secret to enter, so select the dynamic discovery option if Copilot Studio offers one.
Select Create, then select Next.
Select Create a new connection, sign in with your Vetera details, then select Add to agent.
After the agent is published to Microsoft 365 Copilot, users can pin it and ask questions about your Vetera data from chat.
Using a Configuration File (For Developer Tools)
Use this option for developer-style tools such as Claude Code, Cursor, or Codex.
Add the server to your MCP configuration under mcpServers. The example below uses the endpoint for Germany, Austria, and Switzerland. If your Open API access settings show a different endpoint URL, use that one.
{
"mcpServers": {
"vetera": {
"type": "http",
"url": "https://mcp.provetcloud.app/mcp"
}
}
}
Finish the Connection
Follow these steps after connecting with Claude, ChatGPT, Gemini, or a configuration file.
When a browser window opens, paste any URL from your current Vetera session. This tells the connector which clinic and organization you belong to.
Select Authorize.
Access is granted automatically if you're eligible for Vetera MCP.
Suggested Prompts
You do not need a skill to start using Vetera MCP. Once your AI tool is connected, you can ask questions directly. These prompts are a good starting point.
Income and services
"How did we do last month?"
"Show me our income for each of the last 12 months."
"Which of our services bring in the most money?"
"Compare this month to the same month last year."
"Which of our locations is doing best?"
Money owed
"How much money are we owed right now?"
"Who are our top 10 clients by amount owed?"
"How much has been owed for more than 90 days?"
"How much are we waiting on from insurance companies?"
"Give me a chase list: everyone owing more than €200 for over 60 days."
"On average, how long do clients take to pay us?"
Appointments and diary
"What are our busiest days of the week?"
"How many no-shows did we have last month?"
"What are the most common reasons appointments get cancelled?"
"How long do clients wait between arriving and being seen?"
"How many appointments does each vet have this month?"
"Compare our appointment numbers this quarter to last quarter."
Clients and patients
"How many new clients did we get each month this year?"
"Which patients haven't been seen in over a year?"
"What's our mix of dogs, cats, and other species?"
Health plans
"How many active health plan subscriptions do we have?"
"Are our health plan members actually using what's included?"
"How many health plans have overdue payments?"
"Which health plans get cancelled most, and why?"
If you also use other connectors, such as accounting software, you can ask questions that span both in a single conversation — for example, comparing clinic revenue against costs.
The starter skills build on these prompts, returning the same consistently structured report each time you ask.
Get and Use Starter Skills
Skills are ready-made instructions you give your AI tool so it always returns the same kind of report for the same question. Each skill is a single file that includes both the analysis your AI tool should run and starter prompts you can copy.
There are five starter skills. To get the skill files, contact Vetera support.
Skill | What it covers | Try asking |
Practice Health Check | Income compared with last month and the same month last year, invoice count, top 5 services, consultation count. | "Run my practice health check for last month." |
Getting Paid | Who owes the most, how old the debt is, insurance vs. client split, average days to payment. | "Who owes us money? Run my debtors check." |
Appointments & Diary | Busiest days and hours, no-show and cancellation patterns, waiting times. | "Review my diary for the last three months." |
Clients, Patients & Health Plans | New clients, lapsed patients for recall campaigns, health plan utilization and overdue payments. | "How's our client base looking this year?" |
Clinic Data Analyst (agent) | Answers questions across all the areas above, for anyone who'd rather not pick a single skill. | "I have 10 minutes. Give me the state of the practice." |
Load a Skill Into Your AI Tool
Claude
Add it to a Project: select Projects, then create a new project. Paste the skill file's contents into the project's custom instructions, or upload the file to the project's knowledge. Every chat in that project then follows the skill.
Use the agent file the same way, in its own dedicated project (for example, "Clinic Analyst").
ChatGPT
Create a Project and paste the skill file's contents into its instructions, or attach the file and ask ChatGPT to follow it.
Or build a Custom GPT (Explore GPTs > Create > Configure) with the skill's contents as the instructions.
Gemini
Open the Gem manager, select New Gem, paste the skill's contents into the Gem's instructions, save, then chat with that Gem.
We recommend one skill per project, GPT, or Gem, plus one general assistant built from the agent file.
Example Use Cases
Monthly Business Review
For a clinic owner, on the first Monday of the month.
Ask: "Run my practice health check for last month."
You get: income compared with the previous month and the same month last year, invoice count, top 5 services, and consultation count.
Reducing Money Owed
For a practice manager, as a weekly routine.
Ask: "Give me this week's chase list."
You get: who owes the most, how old the debt is, the insurance vs. client split, and the average days to payment.
Staffing the Diary
For a manager planning rotas.
Ask: "When are we busiest, and where are we losing slots?"
You get: busiest days and hours, and no-show and cancellation patterns.
Filling Quiet Weeks With Recalls
For reception or nursing teams.
Ask: "Find patients we haven't seen in over a year."
You get: a lapsed-patient list to drive reminder and vaccination-recall campaigns.
Quarterly Health Plan Check
For a clinic owner.
Ask: "Review our health plans — are they working?"
You get: active subscriptions, utilization, overdue payments, and top cancellation reasons.
Combining Data Sources
For clinics that also use other connectors, such as accounting software.
Ask a cross-system question in one conversation — for example, comparing clinic revenue against costs.
FAQs
Which AI Tools Work With Vetera MCP?
Any MCP-compatible client, including Claude, ChatGPT, Gemini, and Microsoft Copilot.
Is Vetera MCP Read-Only?
Yes. Vetera MCP cannot make changes to your Vetera data.
Who Can Connect?
Anyone with the Administrator role, membership in your organization's Vetera MCP personnel group, or a shared department with that group.
I Can't Find the Option to Add a Custom Connector in Claude
If your organization uses a Claude Team or Enterprise plan, only your Claude account owner can add a new custom connector. Ask them to complete the setup step described in Connect Your AI Tool above. Once they've added it, you can connect to it yourself.
I See an Error Saying My Clinic Is Not Configured for MCP
Vetera MCP has not been enabled for your clinic yet. A clinic administrator needs to add it in Settings > General > Integrations > Open API access. Contact your clinic administrator.
I Have Access but Still Cannot Connect
Only the built-in Administrator role has access by default. A custom administrator role your organization created does not. Ask your clinic administrator to add you to the Vetera MCP personnel group.
Does Vetera MCP Include Historical Migrated Data?
No, not in the short to medium term.
Why Might the Numbers Differ From a Report I Already Use?
Vetera MCP includes only finalized invoices, and includes credit notes, matching how Vetera's own reports calculate totals.