DKG Node Services

Learn how to navigate and use the DKG Node’s built-in web interfaces and APIs. This section helps you manage your node, interact with agents, and explore services without writing complex code.

ning your DKG Node in development mode

You will be running your DKG Node in development mode while building, experimenting, and customizing your DKG Node, before deploying it in production. In this mode, the system automatically reloads on code changes, streams real-time logs, and gives you immediate feedback as you work.

From the project root:

cd ~/dkg-node && npm run dev

This will:

  • Automatically reload your node whenever you change the code.

  • Stream live logs across all running services.

  • Help you debug and iterate quickly in a local environment.

Troubleshooting

If npm install fails, try:

rm -rf node_modules package-lock.json
npm install

Also confirm your Node.js version is v22+.

Local services & dashboards

Once your dev server is up (npm run dev), several powerful tools become available through your browser. These interfaces let you manage, inspect, and debug every part of your DKG Node.

DKG Node & Agent UI

http://localhost:8081/

This is where the prebuilt template UI is for your DKG Node. From here, you can:

  • Monitor the overall health and status of your node.

  • View and manage installed plugins.

  • Interact directly with your DKG Agent (the built-in AI agent).

  • Access settings, credentials, and configuration options.

  • Publish, query, and verify Knowledge Assets through a graphical interface - no code required.

Think of this as your command center for operating and experimenting with your node.

DKG MCP Server

http://localhost:9200/mcp

This is the Model Context Protocol (MCP) server endpoint your node exposes. It allows:

  • AI agents and external applications to connect to your node.

  • Execution of tools and APIs provided by installed plugins.

  • Structured communication between your node and LLMs or external services.

If your DKG Node is the “brain,” the MCP server is the communication layer - it’s what lets AI systems talk to your node programmatically.

Swagger UI (API Explorer)

http://localhost:9200/swagger

The Swagger dashboard provides interactive documentation for all the REST APIs your node exposes. Here, you can:

  • Explore every available endpoint.

  • Understand request/response formats.

  • Test API calls directly from the browser.

This is especially helpful if you’re integrating the DKG Node into a larger application or developing custom tools.

Operational database viewer (Drizzle Studio)

https://local.drizzle.studio

Drizzle Studio is a visual interface for inspecting the internal databases your DKG Node uses. It allows you to:

  • Browse tables and view stored data.

  • Inspect Knowledge Assets before and after they’re published.

  • Debug database writes and schema changes.

Last updated

Was this helpful?