MCP

About MCP

MCP (Model Context Protocol) is a protocol that enables AI systems to connect and interact with external tools, software, and data sources in a standardized way, similar to the API in regular software.

In the context of Timebook, you can use it to connect the AI tool of your choice and use it to perform a series of tasks driven by Timebook data. In this article, we cover configuration for Claude, Cursor, and Gemini CLI.

A conversation in Claude based on Timebook data

Tools

Timebook MCP supports the following tools (calls) that you can run within your AI service. In other words, this is the range of activities that you can perform using the Timebook data by asking the AI client of your choice.

Items
Insights
Opportunities
Goals
Contacts
Custom
Projects
Other

Search items by title

Search insights

Search opportunities by query, parameters, or ID

Get current goals

Search contacts, personas, or companies by workplace ID

Get item custom statuses

Get project info

Search teams

Get linked items

Get opportunities linked to insight id

Get insights linked to opportunity id

Get goal key results

Get contact, persona, or company by ID

Get custom fields

Get item projects

Retrieve product info

Update item date range

Link insight to opportunity

Link opportunity to goal

Update goal key result

Get contacts by persona ID

Update item custom field

Create project

Create discovery board

Link task to item

Get basic opportunities info

Update item custom status

Assign opportunity to project

Search document by ID

Delete linked item

Search documents by workplace ID

Configuration

Please mind that, for now, all MCP integrations require Google as Timebook's authorization method.

Claude Desktop

Claude Desktop is a free edition of Claude which does not have an MCP client in app. Instead, it uses mcp-remote, a Node.js module launched with npx. Make sure to download and install Node.js before proceeding.

1

Go to Settings → Developer and click 'Edit Config'

MCP settings location in Claude
2

This will take you to the claude_desktop_config.json location in your OS. Open the file in a text editor and paste the following:

{
  "mcpServers": {
    "timebook-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://mcp.timebook.net/mcp/"
      ]
    }
  }
}

The configuration allows Claude Desktop to connect to Timebook and acquire an access token required to register itself as an MCP client in the service.

3

Save the changes and restart Claude Desktop.

4

Once you launch it again, you will be moved to the Timebook sign-in page. Sign in with your Google account to finish configuration.

If you configured everything correctly, Timebook will be listed as an MCP server in the developer settings:

A list of connected MCP servers in Claude

Cursor

1

Go to Cursor settingsTools & MCP and click 'Add Custom MCP'.

2

This will take you to the Cursor's mcp.json file. Copy the following and paste it into the file:

{
  "mcpServers": {
    "timebook-mcp": {
      "url": "https://mcp.timebook.net/mcp/",
      "description": "Timebook MCP server serving Insights and Opportunities",
      "transport": {
        "type": "http-streamable"
      }
    }
  }
}
3

The Timebook MCP server will appear on the list. Click 'Connect' to continue.

A list of connected MCP servers in Cursor
4

Proceed with the instructions on screen to authenticate Cursor in Timebook.

If you configured everything correctly, Timebook will appear as an enabled server on the list.

Timebook MCP enabled in Cursor

Gemini CLI

1

Go to ~/.gemini/settings.json and add an MCP server object with the following configuration:

  "mcpServers": {
    "timebook-mcp-server": {
      "httpUrl": "https://mcp.timebook.net/mcp/",
      "authProviderType": "dynamic_discovery"
    }
  }
2

Launch the CLI by running gemini in the terminal. On the first run, it will ask you to approve its authorization in Timebook. Proceed with the instructions to finish configuration.

Timebook asking for authorization consent for the Gemini CLI

Last updated

Was this helpful?