GoodTime MCP Server

Last updated: June 30, 2026

The GoodTime MCP Server allows AI assistants like Claude to securely access your live GoodTime data using natural language. Instead of manually exporting reports or learning our API, you can simply ask Claude questions such as:

  • Show me all interviews scheduled next week.

  • Which interviewers have conducted the most interviews this month?

  • Analyze interview load and identify potential interviewer burnout.

  • Compare completed interview volume over the last 90 days.

Claude will automatically translate your request into the appropriate GoodTime API calls and return a summarized answer.

Coming Soon: One-Click Installation

We’re excited to announce that the GoodTime MCP Server will soon be available directly through the Claude App Directory, making setup as simple as a few clicks.

Until then, you can manually configure the GoodTime MCP Server by following the steps below.

Please NOTE: The GoodTime MCP Server has currently been built and tested with Claude AI; however we do have longer term plans to make this MCP connection available to other AI tools. You are welcome to test this MCP connection with other AI tools, however internal testing has only been conducted with Claude AI, and the following instructions reflect this specific integration.


What is MCP?

MCP (Model Context Protocol) is an open standard that allows AI assistants to securely connect to external applications through structured tools.

Unlike uploading data into an AI model, MCP gives Claude live, secure access to your GoodTime account whenever you ask a question.

The flow looks like this:

You
   ↓
Claude
   ↓
GoodTime MCP Server
   ↓
GoodTime Public API
   ↓
Live GoodTime data

This means Claude is always working from your current recruiting data—not stale exports or spreadsheets.

Learn more about MCP at:
https://modelcontextprotocol.io/docs/getting-started/intro


Before You Begin

You’ll need:

  • A GoodTime account

  • A GoodTime API Access Token (v2)

  • Claude Desktop

  • Node.js 18 or later (required for the manual installation)


Step 1: Install Claude Desktop

Download Claude Desktop from:

https://claude.com/download

Install and launch the application before continuing.


Step 2: Generate a GoodTime API Token

The MCP Server authenticates using your GoodTime API token.

  1. Log into GoodTime.

  2. Navigate to Settings → API Access.

  3. Or visit:

    https://a.goodtime.io/settings/auth/api-access

    (NOTE: If you do not see this menu item in your GoodTime menu, please ask your CSM to activate it for you)

  4. Select Generate New Access Token (v2).

  5. Copy the generated token—you’ll need it in the next step.

Note: Treat your API token like a password. Do not share it publicly.


Step 3: Configure Claude

Locate the Claude Desktop configuration file:

~/Library/Application Support/Claude/claude_desktop_config.json

Add the following configuration:

{
  "mcpServers": {
    "goodtime-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "goodtime-mcp"
      ],
      "env": {
        "GOODTIME_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Replace:

YOUR_API_KEY

with the API token you generated in the previous step.

Save the file.


Step 4: Restart Claude

Completely quit Claude Desktop and reopen it.

Claude will automatically load the GoodTime MCP Server during startup.


Verify Your Connection

To confirm everything is working, try asking Claude:

Show me my upcoming GoodTime interviews for this week.

Or open Claude’s MCP tool list using:

/mcp

If the GoodTime MCP Server appears, your installation is complete.


What Can You Ask Claude?

Once connected, Claude can answer questions using your live GoodTime recruiting data.

For example:

Interview Activity

  • Show all interviews scheduled next week.

  • List completed interviews from last month.

  • Compare interview volume over the last two quarters.

Interviewer Insights

  • Who conducted the most interviews this month?

  • Find active interviewers in Engineering.

  • Which interviewers haven’t participated in an interview recently?

Capacity Planning

  • Which interviewer pools are nearing capacity?

  • If interview volume increased by 50%, where would bottlenecks occur?

Operational Analysis

  • Analyze interviewer workload.

  • Identify potential interview burnout.

  • Summarize recruiting activity by team.


Current Capabilities

Version 1.0 of the GoodTime MCP Server provides read-only access to GoodTime data.

Supported data includes:

  • Interviews

  • Contacts (Interviewers, Recruiters, Coordinators)

  • Interviewer Tags

  • Tag Categories

Write operations—such as scheduling interviews, creating candidates, or updating interview records—are not currently supported.


Troubleshooting

If Claude cannot access your GoodTime data, check the following:

Invalid API Token

Ensure you’re using a valid GoodTime API Access Token (v2).

Missing Permissions

Your GoodTime user account must have permission to access the requested data.

Restart Claude

Claude only loads MCP servers during startup. Restart the application after making configuration changes.

Node.js

The manual installation requires Node.js 18 or later to be installed.


Additional Resources

GoodTime MCP Package (npm)

https://www.npmjs.com/package/goodtime-mcp

Model Context Protocol Documentation

https://modelcontextprotocol.io/docs/getting-started/intro