Connect to Lansweeper's MCP server

Connecting an MCP-compatible AI assistant to Lansweeper is a one-time setup per device or workspace. Once connected, your assistant can query your IT estate in plain language. For background on what the Lansweeper MCP server is and what you can ask, see Explore the Lansweeper MCP server.

Prerequisites

Before you connect:

• A Lansweeper account with access to at least one site
• An MCP-compatible AI assistant

Connect your AI assistant

Lansweeper's MCP server supports the following clients: Claude Desktop, Claude Code, Cursor, Atlassian Rovo, Copilot Studio, ChatGPT, and Gemini. For most, setup follows the same pattern: add the endpoint URL and sign in with your Lansweeper credentials when prompted. Steps specific to each client are listed below.

Claude Desktop

1. Open Claude Desktop and select Customize.

2. Select Connectors, then Add custom connector.

3. Enter a name and the endpoint URL for your region, then select Add.

4. Authenticate with your Lansweeper credentials when prompted.

If you prefer to configure manually, you can edit the configuration file directly:

• Windows: %APPDATA%\Claude\claude_desktop_config.json

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

  Add the Lansweeper MCP server entry to the mcpServers object. Add both entries if you have sites in both regions:

  {
   "mcpServers": {
     "lansweeper": {
       "type": "http",
       "url": "https://mcp.lansweeper.com/mcp"
     }
   }
  }
  

  Save the file and restart Claude Desktop. On your first query, a browser window opens for sign-in.

Claude Code

1. Open a terminal and run the following commands to register the Lansweeper MCP server. Run both if you have sites in both regions:

   claude mcp add --scope user --transport http lansweeper https://mcp.lansweeper.com/mcp
   

2. In Claude Code, type /mcp and select lansweeper from the list.

3. Select Authenticate. A browser opens. Sign in with your Lansweeper credentials.

4. Once the browser shows a Connected confirmation, return to Claude Code. The server is ready.

Cursor

1. Open or create Cursor's MCP configuration file at ~/.cursor/mcp.json.

2. Add the Lansweeper MCP server entry. Add both entries if you have sites in both regions:

   {
    "mcpServers": {
      "lansweeper": {
        "type": "http",
        "url": "https://mcp.lansweeper.com/mcp"
      }
    }
   }
   

3. Save the file and restart Cursor.

4. On your first query, a browser window opens for sign-in. Sign in with your Lansweeper credentials.

Atlassian Rovo

1. In Rovo, open Connected Apps.

2. Select Add external MCP server.

3. Enter a name (for example, Lansweeper) and the endpoint URL: https://mcp.lansweeper.com/mcp. Select HTTP as the transport type.

4. Save. Rovo authenticates via SSO, so no separate sign-in is required.

Copilot Studio

1. Go to copilotstudio.microsoft.com and sign in.

2. Select Create > New agent and give the agent a name.

3. Go to Settings > Generative AI > Orchestration, set it to Yes, then select Save.

4. Select the Tools tab, then select Add a tool > New tool > Model Context Protocol.

5. Enter a name (for example, Lansweeper) and the endpoint URL: https://mcp.lansweeper.com/mcp.

6. Set authentication to OAuth 2.0 > Dynamic discovery.

7. Select Create, then Next.

8. Select Add and configure.

9. Authorize the connection by signing in with your Lansweeper account when prompted.

10. Test the connection from within the agent.

ChatGPT

1. Open ChatGPT and select your profile icon in the bottom-left corner.

2. Go to Settings > Apps and Connectors > Advanced Settings and enable Developer Mode.

3. Go back to Apps and Connectors and select Create.

4. Enter a name (for example, Lansweeper) and the endpoint URL: https://mcp.lansweeper.com/mcp.

5. Select Save. To make the connector available to others in your workspace, publish it.

6. Start a new chat, select + in the message bar, then More, and select the Lansweeper connector.

Gemini

You can connect Lansweeper to Gemini in two ways: through Gemini Enterprise for shared, company-wide use, or through the Gemini CLI for individual local use. Choose the one that matches how you work.

Gemini Enterprise

Use this option to make Lansweeper available to a shared Gemini Enterprise agent that everyone in your company can query. Setup happens in the Google Cloud console and needs a Google Workspace or Cloud admin.

Before you start:

• Your admin account needs the Discovery Engine Editor role in Google Cloud IAM. Without it, data store creation fails.
• If your organization uses policy restrictions or VPC Service Controls, allowlist mcp.lansweeper.com, mcp.auth.lansweeper.com, and eu.lansweeper.com first. Custom MCP Server is a Preview feature and can be blocked by default.

1. In the Google Cloud console, go to Gemini Enterprise > Data stores > Create data store.

2. Search for Custom MCP Server (Preview), then select Add MCP server.

3. Generate a Client ID and Client secret in Lansweeper at MCP OAuth clients.

4. When you create the Lansweeper OAuth client, add Google's redirect URI, https://vertexaisearch.cloud.google.com/oauth-redirect, as an authorized redirect URI. If you skip this, Verify Auth fails in step 6.

5. Back in the Google Cloud console, select OAuth 2.0 and enter:

   • MCP Server URL: https://mcp.lansweeper.com/mcp
   • Authorization URL: https://mcp.auth.lansweeper.com/oauth2/authorize
   • Token URL: https://mcp.auth.lansweeper.com/oauth2/token
   • The Client ID and Client secret from step 3
   • Scopes: openid, offline_access, email, phone, profile

   These OAuth values come from Lansweeper's discovery document at https://mcp.lansweeper.com/.well-known/oauth-authorization-server. Check there if the endpoints ever change.

6. Select Verify Auth, then Continue.

7. Enter a description of what the server does. The agent reads this to decide when to call Lansweeper, so be specific. For example: Query Lansweeper IT asset inventory: devices, software, vulnerabilities, users.

8. Pick a region and a name for the data store, then select Create. Wait for the status to change from Creating to Active.

9. Open the data store and go to Actions. A new data store has all tools disabled. Select Reload custom actions, then enable the ones you want available (up to 100).

10. Open your Gemini Enterprise agent, add the data store as a connected data source, and test it with a real question. The data store does nothing until it's connected to an agent or app.

Gemini CLI

Use this option for local development with the Gemini CLI. No Google Cloud console is needed.

1. Open or create ~/.gemini/settings.json.

2. Add the Lansweeper MCP server under mcpServers:

   {
     "mcpServers": {
       "lansweeper": {
         "httpUrl": "https://mcp.lansweeper.com/mcp"
       }
     }
   }
   

3. Start gemini and ask something that needs Lansweeper data, or run /mcp auth lansweeper. A browser window opens for you to sign in and authorize. The Gemini CLI auto-discovers the OAuth endpoints, so a Client ID or secret usually isn't required.

4. If auto-discovery fails, generate a Client ID and secret as in the Gemini Enterprise section (step 3), then add them to settings.json:

   {
     "mcpServers": {
       "lansweeper": {
         "httpUrl": "https://mcp.lansweeper.com/mcp",
         "oauth": {
           "enabled": true,
           "clientId": "your-client-id",
           "clientSecret": "your-client-secret",
           "authorizationUrl": "https://mcp.auth.lansweeper.com/oauth2/authorize",
           "tokenUrl": "https://mcp.auth.lansweeper.com/oauth2/token",
           "scopes": ["openid", "offline_access", "email", "phone", "profile"]
         }
       }
     }
   }
   

The CLI stores tokens locally in ~/.gemini/mcp-oauth-tokens.json and refreshes them automatically. After the first sign-in, no further action is needed.

Limits

Troubleshooting

Next steps

• Explore the Lansweeper MCP server
• Data domains and tools