The MCP Server lets AI assistants and AI agents, such as ChatGPT or Microsoft Copilot, securely access your Meisterplan data and work directly with your portfolio and project data from your AI application.
- Background: What Is an MCP Server?
- Prerequisites
- What Our MCP Server Can Do
- Frequently Asked Questions
Background: What Is an MCP Server?
The Meisterplan MCP Server (Model Context Protocol) is a standardized interface that lets AI agents securely access Meisterplan data, without having to export and manually transfer it to your AI agent.
Similar to the REST API, which makes Meisterplan data available to other applications, the MCP Server makes this data available specifically to AI agents. When you ask a question, the AI agent automatically knows which Meisterplan data it needs.
The quality of the answers depends on the AI model you use. More capable models interact more effectively with the MCP Server and deliver more precise results.
Prerequisites
Before connecting an MCP Server, make sure your LLM provider supports MCP and that such connections are permitted by your organization's security policies.
Required User Rights
To create an API token for the MCP Server, you need to have the Access MCP Server right enabled under Manage > User Groups > Views & Areas > AI Functionality.
Creating an API Token for the MCP Server
Click your profile picture in the top right and select My Profile. In the toolbar, click Manage Apps > API Tokens > Add API Token and select MCP Server:
Click the Copy to Clipboard icon to copy the API Token and store it in a secure location. After closing the dialog, you will no longer be able to view this value.
User Account
Which data you can retrieve via the MCP Server depends on the rights of the user account used to establish the connection. The MCP Server respects all field-level permissions of that user account.
For central setups, we recommend creating a dedicated user account that is used exclusively for the MCP Server. This ensures that all users of a shared AI agent see exactly the data they are intended to see, regardless of their own rights in Meisterplan.
Create this account under Manage > Users and add it to a user group whose rights match exactly what all users of the central agent should be able to see.
What Our MCP Server Can Do
With our MCP Server, you can get answers to portfolio and project questions without switching between Meisterplan and your AI agent. You can also use Meisterplan data in other workflows and reports.
You can currently retrieve the following Meisterplan data via the MCP Server, provided you have at least read-only rights:
- Projects with project details (e.g. Status, Goals including configured colors, Rank)
- Project KPIs (e.g., Total Costs, Total Effort, Total Capex, Total Opex, Total Benefits, Net Value)
- Scenarios
- Project Comments
- Allocation Comments
- Portfolios
If a request includes data you do not have access to, the AI agent will return a partial answer along with a hint that some data was not accessible.
Use Cases
The following examples show how you can use the MCP Server in your day-to-day work:
Keeping Track of Critical Projects
Your portfolio contains dozens of projects and you want to quickly know which ones are critical and why, without opening each project individually.
Ask your AI agent:
| List all projects with the status Critical and explain why they might be critical. |
The AI agent identifies the affected projects and derives possible reasons from the associated project fields and KPIs.
Using Similar Projects for Planning
You are planning a new project and want to benefit from the experience of similar projects, but don't have time to search the portfolio manually.
Ask your AI agent:
| I want to create a new project X. Select three similar projects from the portfolio and create an overview of their key project fields and values. |
The AI agent finds comparable projects and compiles the relevant key figures so you can make realistic assumptions for your new plan.
Preparing for Meetings
You have a status meeting tomorrow and want to be up to date in a few minutes instead of having to tediously gather information from different views.
Ask your AI agent:
| I have a status meeting on project XY tomorrow. Summarize the most important key figures and current information for me. |
The AI agent compiles an overview of status, schedule, financial key figures, and relevant project fields, so you go into the meeting prepared.
Note on Costs:
AI agents typically charge for the amount of data processed per token. For large portfolios with many projects and custom fields, individual requests can consume a significant number of tokens. We recommend narrowing your questions to relevant projects or fields.
Frequently Asked Questions
Question: Where is the MCP Server hosted?
Answer: The Meisterplan MCP Server is hosted in the configured AWS region, either in the United States (Oregon) or within the European Union.
AWS documents that user inputs and the model outputs are not shared with the model providers (source).
---
Question: Is customer data used for AI? What data?
Answer: When using the MCP Server, the retrieved project data is transmitted to the AI agent that your Meisterplan administration has connected to the MCP Server. The data that is transmitted depends on the rights of the connected user account and the request being made.
---
Question: Will the data be used to train or update the AI?
Answer: The data retrieved via the MCP Server is transmitted to the AI agent of your chosen provider. Whether and how that provider processes the data further is governed by their terms of use. Meisterplan has no influence over this.