DillyDallyMCP

You run a Model Context Protocol (MCP) server that integrates with Dedalus to manage and expose your DillyDally data endpoints. This server lets MCP clients request activity, productivity, sessions, and other metrics through a consistent interface, enabling rapid testing and deployment in local and cloud environments.

How to use

Start the server in two modes to suit your testing or integration needs. In stdio mode, you run the MCP server locally and connect clients directly. In http mode, you expose the server over HTTP for remote clients to consume.

To run in stdio mode, use the standard npm script that starts the MCP server locally. To run in http mode, use the HTTP-oriented script that starts a local HTTP endpoint for debugging and testing.

How to install

Prerequisites: you need Node.js installed on your system. You should also have npm available to install packages and run scripts.

# 1) Install dependencies
npm install

# 2) Build the project
npm run build

# 3) Start in stdio mode (MCP clients)
npm run dev:stdio

# 4) Start in HTTP mode (testing/debugging)
npm run dev:http

When you start in stdio mode, the MCP server runs locally and accepts connections from MCP clients via the stdio transport. When you start in HTTP mode, the server exposes an HTTP endpoint at a local address for testing.

Additional notes

Environment variables are configured in a local file named .env.local within the project. The key variable shown is CONVEX_URL, which should point to your Convex deployment, e.g. https://your-deployment.convex.cloud.

The local .env.local file is gitignored to keep sensitive configuration out of version control. Create it to point the server to your deployment, then keep it local to your development environment.

Available tools

add_integers

Adds two integers together and returns the sum.

get_recent_activity

Fetches recent activity snapshots from DillyDally to surface the latest interactions and events.

get_last_session

Retrieves details about the most recent DillyDally session, including participants and duration.

get_productivity_stats

Calculates productivity statistics over a specified time range to help you measure focus and output.

get_session_details

Provides detailed information about a specific DillyDally session.

get_attention_metrics

Returns attention or focus metrics derived from camera snapshots.