Connecting Claude to the DAZZM MCP server
Updated on Published on
Article summary
Connect Claude to a DAZZM application through the MCP server. Generate an API key in Account, add the connector from Claude's settings — or through the configuration file — then let Claude explore and modify your application: definitions, pages, workflows and data.
Introduction
Section titled “Introduction”The DAZZM MCP (Model Context Protocol) server lets you connect Claude directly to a DAZZM application. Once the connection is established, Claude can act in two areas:
- Modify the application by manipulating its definitions (models, pages, workflows, scripts, etc.);
- Read and modify the data contained in the application.
There are two ways to connect Claude to a DAZZM application:
- From Claude’s interface. You add the DAZZM server as a custom connector directly from Claude’s settings. Nothing to install, no file to edit. This works the same way in the Claude desktop app and in the browser (claude.ai), and the connector is tied to your Claude account: it follows you if you switch computers;
- Through the configuration file (
claude_desktop_config.json). A more technical approach, useful if you prefer managing your connectors as code or if you already maintain that file. This method is local to Claude Desktop on a given machine.
Both methods are covered below, along with Claude Code CLI, the command-line version. The preparation steps — generating the API key, choosing the environment, usage tips — are identical in every case.
N.B.: The MCP server was designed to be connected to a development environment, where modifying definitions is allowed. See the Choosing the right environment section before connecting to any other environment.
Prerequisites
Section titled “Prerequisites”- Access to your DAZZM organization’s Account application;
- A DAZZM application with a development environment;
- Claude (web or desktop application) or Claude Code CLI;
- Node.js installed — only if you use the configuration file method (it relies on
npx).
Steps to follow
Section titled “Steps to follow”1. Generate an API key (Account)
Section titled “1. Generate an API key (Account)”- Open the Account application;
- Access the application you want to connect;
- Access a development environment of that application;
- Create an API key. You can name it, for example, “Claude”;
- Copy the key’s value.
2. Add the connector from Claude’s interface
Section titled “2. Add the connector from Claude’s interface”With this method, the connector is added to your Claude account, so once set up it becomes available wherever you use Claude — desktop app or browser, Claude Chat, Claude Cowork and Claude Code — and on any computer you sign in on.
-
In Claude, open Settings, then the Connectors section;
-
Click Add custom connector;
-
Fill in the two fields:
- Name: the name the connector will have in your connectors list. We suggest
DAZZMfollowed by your application’s name — particularly useful if you control several applications through Claude; - URL:
https://<app-name>-<env-name>.apps.dazzm.com/api/prod/mcp, where<app-name>is the name of your application and<env-name>is the name of the environment you are connecting to — the same URL segment you use to access your application.
N.B.: The
prodsegment in/api/prod/mcpis fixed: it refers to the version of the DAZZM platform, not to your application’s environment. Do not replace it. - Name: the name the connector will have in your connectors list. We suggest
-
Click Continue. Claude tries to check the server and displays the message “This URL can’t be checked automatically. You can still configure it manually.” This is expected: the DAZZM server requires an API key, so Claude cannot reach it before it is configured. Click Continue anyway;
-
On the next screen, in the Authentication section, select No sign-in;
-
Still in that section, add a request header:
- Header name:
api-key - Value: the key value copied in step 1.
- Header name:
-
Click Connect.
The connector now appears in your connectors list, ready to use. Move on to Verify the connection.
3. Configure Claude through the configuration file (alternative)
Section titled “3. Configure Claude through the configuration file (alternative)”This method is an alternative to step 2 — you do not need both. It applies to all of Claude Desktop: Claude Chat, Claude Cowork and Claude Code will all have access to it. (For the command-line version of Claude Code, see step 4.)
Edit the Claude Desktop configuration file:
Claude\claude_desktop_config.jsonLocate the mcpServers section. It may not exist by default; in that case, create it:
{ "mcpServers": {}}Then add an entry for your application. Here is a complete example:
{ "mcpServers": { "dazzm-<app-name>": { "command": "npx", "args": [ "-y", "mcp-remote", "https://<app-name>-<env-name>.apps.dazzm.com/api/prod/mcp", "--header", "api-key:<YOUR_API_KEY>" ] } }}A few notes about this example:
dazzm-<app-name>is the name of the MCP server as Claude will see it. We recommend naming itdazzm-followed by your application’s name. This is particularly useful if you control several applications through Claude;- The URL follows the format
https://<app-name>-<env-name>.apps.dazzm.com/api/prod/mcp, where<app-name>is the name of your application and<env-name>is the name of the environment you are connecting to — the same URL segment you use to access your application; - The
prodsegment in/api/prod/mcpis fixed: it refers to the version of the DAZZM platform, not to your application’s environment. Do not replace it; api-key:is followed directly by the key value copied in step 1, with no space.
Save the file, then quit Claude and relaunch it.
N.B.: On Windows, you must quit Claude from the system tray (notification area), not just close its window.
4. Configure Claude Code CLI
Section titled “4. Configure Claude Code CLI”Coming soon: the configuration procedure for Claude Code CLI. The preparation (API key, URL) is identical to Claude Desktop’s; only the place where the configuration is stored differs.
5. Verify the connection
Section titled “5. Verify the connection”Start a conversation in Claude Chat, Claude Cowork or Claude Code and press the + button: the list of active connectors should include your DAZZM server.
If you used the configuration file, Claude connects to the configured MCP servers when it launches. If one of them does not work, an error is displayed immediately. If there is no error, the MCP server most likely loaded successfully.
If you connect Claude to a brand-new application, it has few examples to rely on to find the right solutions. The first two tips help it find some:
Connect a reference application. Register a second MCP connection, this time to one of your complete, well-built applications. Choose any application you like, create an API key for it as well (step 1), then add it the same way and name that connector dazzm-examples. The name matters: it is what tells Claude that this is a registry of examples, and Claude will intuitively use it to draw on good DAZZM practices.
Point Claude to the documentation. At the start of a conversation, you can simply tell Claude to visit doc.dazzm.com to find more examples and guides about the platform.
Let Claude test in the application. To make sure Claude understands and tests what it is doing, it is really useful to give it access to the application it is working on. This is easy: open your browser already logged into the application and tell Claude to use the Chrome MCP to run tests on the application it is modifying. A good practice is to ask it to run end-to-end tests every time it has completed a task.
Choosing the right environment
Section titled “Choosing the right environment”The MCP server can act on any environment, but not all uses are equal:
- Nominal case — development environment. This is what the MCP server was designed for: Claude can freely manipulate definitions and test data there.
- Acceptable case — production environment, read-only. It is possible to connect the MCP server to a production environment if the goal is to use its data reading capabilities for analysis.
- Strongly discouraged case — writing data outside a test environment.