Prerequisites
The MCP integration is in preview - tools, parameters, and behavior may change between releases.
Before you connect an MCP client to Dime.Scheduler, make sure the following pieces are in place.
A Dime.Scheduler cloud subscription
The MCP server is hosted as part of the Dime.Scheduler cloud platform. You need an active subscription on the sandbox environment to try it out today. See Configuration for endpoint URLs and rollout status.
An API key
The MCP server authenticates with the same API keys used by the public API. Every request must include an X-API-KEY header.
To create one:
- Sign in to Dime.Scheduler.
- Go to Administration → API Keys.
- Create a new key, give it a meaningful name (e.g.
MCP – my-laptop), and copy the value. - Store it in your client's secret store - never commit it to a repository or paste it into chat.
Create a dedicated API key per MCP client (one for Claude Desktop, one for Cursor, one for CI, …) so you can rotate or revoke them independently if a device is lost.
An MCP-capable client
Any client that speaks the Model Context Protocol over Streamable HTTP can connect. Tested clients include:
- Claude Desktop
- Claude Code
- Cursor
- MCP Inspector - recommended for first-time exploration
- Custom clients built with the official TypeScript or Python SDKs
If your client only supports the older stdio transport, you can bridge to the remote endpoint with mcp-remote - see Clients for examples.
Network access
The endpoint is a regular HTTPS URL. There is no special firewall configuration required as long as your client can reach *.api.dimescheduler.com on port 443. If your organization terminates outbound TLS, allow-list the host.
Familiarity with your data
The tools operate on the names and identifiers that already exist in your Dime.Scheduler tenant - resource display names, task numbers, category names, and so on. The richer your master data, the better the assistant will be at finding the right records on the first try.