AggreGate

AggreGate MCP Server provides a standardized interface for AI assistants to interact with an AggreGate server. It exposes a rich set of MCP tools to manage contexts, variables, functions, events, devices, users, agents, widgets and dashboards, as well as various model types (relative, absolute and instance-based). The server acts as a bridge, translating MCP requests from AI agents into concrete AggreGate API calls and returning results in a consistent format. Typical workflows include describing variables, creating and binding models, triggering events, querying device statuses, and executing Expression/Query functions within AggreGate contexts. To start using it, configure the MCP server in your client (Cursor in this example) to point at the bundled JAR, then interact with the provided MCP functions through your AI prompts or automation rules. The documentation linked in the project contains detailed guides for workflows, model types, and AggreGate API usage.

Prerequisites:

• Java 11+ installed on your machine
• AggreGate Server available and reachable from the MCP host
• Maven/Gradle wrapper (for building the server from source) if you plan to compile locally

Installation steps:

1. Build the MCP server (if you have the source):
   • Run: ./gradlew build
   • Resulting JAR should be available at build/libs/aggregate-mcp-server-<version>.jar
2. Prepare your MCP client configuration (Cursor example):
   • Ensure Cursor is installed and can read ~/.cursor/mcp.json
   • Create or edit ~/.cursor/mcp.json with: { "mcpServers": { "aggregate": { "command": "java", "args": ["-jar", "C:\mcp-server\build\libs\aggregate-mcp-server-1.0.0.jar"] } } }
3. Start the MCP server JAR if not already running as part of Cursor startup, or run the JAR directly:
   • Command: java -jar C:\mcp-server\build\libs\aggregate-mcp-server-1.0.0.jar
4. Restart Cursor to apply the new MCP configuration and verify connectivity in the UI or logs.

Note: If you are using a docker or other deployment method, adapt the command and jar path accordingly.

Tips and common issues:

• Ensure the AggreGate Server address/credentials are correctly configured in the MCP server or via environment variables when required by your deployment.
• If you encounter connectivity errors, verify network access between the MCP host and the AggreGate server, and check firewall rules.
• When upgrading, keep the MCP JAR path consistent and update the mcp.json configuration accordingly.
• Refer to the docs/MCP_AGGREGATE_CONTEXTS_GUIDE.md for critical context management guidance (devices, models, alerts, widgets, etc.).