Skip to main content
Beta Feature — Remote MCP Servers (used by this integration) is currently in beta. Core functionality is stable, but some behaviors and interfaces may change as we continue development. Contact support if you run into any issues.
Datadog is a monitoring and observability platform. PlayerZero connects to Datadog’s remote MCP server, giving AI agents direct access to your Datadog data during investigations.

​
What You Can Do

Once connected, PlayerZero AI agents can:
  • Query monitors — Check monitor status, search for triggered alerts, and investigate monitoring gaps
  • Search logs — Find relevant log entries across your Datadog environment when debugging issues
  • Retrieve dashboards — Pull dashboard data and metrics to understand system behavior during incidents
  • Investigate traces — Access APM trace data for end-to-end request analysis
  • Cross-reference with code — Combine Datadog observability data with codebase context for faster root cause analysis
The specific tools available depend on the permissions granted to the Datadog service account you use for authentication. See the Datadog prerequisites below for details.

​
Prerequisites

Before connecting Datadog to PlayerZero, you need an API key and Application key from your Datadog organization.
  1. Create an API key in Datadog
    • Go to Organization Settings → API Keys
    • Click New Key, give it a name (e.g. PlayerZero), and save it
    • Copy the key value — you’ll need it during setup
  2. Create an Application key in Datadog
    • Go to Organization Settings → Application Keys
    • Click New Key, give it a name (e.g. PlayerZero), and save it
    • Copy the key value — you’ll need it during setup
  3. Assign permissions to the service account
    • Ensure the account has the appropriate read permissions (e.g. mcp_read) for the data you want AI agents to access
    • If you want agents to create or modify resources (such as muting a monitor), grant write permissions (e.g. mcp_write) as well
For full details on creating a Datadog service account with the correct keys and permissions, see Datadog’s MCP Server setup guide.

​
Setup

This connection uses PlayerZero’s Remote MCP Servers feature. Datadog authenticates via custom headers (API key + Application key), not OAuth.
  1. In PlayerZero, navigate to Settings → Ticketing
  2. Navigate to the Remote MCP Servers page and click New Connector
  3. Fill in the connection details:
    • Name — Datadog (or any descriptive name). Connection names must be unique within your organization.
    • Description — Optional. For example: Production Datadog environment
    • MCP Server URL — Enter the Datadog MCP server endpoint for your region. Check Datadog’s documentation for the correct regional URL.
  4. Under Project Access, choose which PlayerZero projects should have access:
    • All projects — Every project in your organization can use this connection
    • Specific projects — Select individual projects
  5. Under Authentication, select Custom
  6. Add two custom headers:
    • Key: DD-API-KEY → Value: your Datadog API key
    • Key: DD-APPLICATION-KEY → Value: your Datadog Application key
  7. Click Add
PlayerZero validates the connection by reaching out to the Datadog MCP server. If the connection is successful, tools from Datadog are automatically available in Player sessions for the selected projects. If the connection fails, PlayerZero will display an error message. Common issues include:
  • Authentication failed — Verify your API key and Application key are correct
  • Connection timed out — Verify the MCP Server URL is correct and the server is reachable
  • Could not resolve host — Verify the MCP Server URL is correct

​
Using Datadog in Player Sessions

After setup, AI agents in Player sessions can access Datadog data alongside your codebase. You don’t need to do anything special — agents automatically use Datadog tools when relevant to the investigation. For example, you can ask a Player: 
Check if there are any triggered monitors related to the payment service.

Search Datadog logs for errors in the checkout API from the last hour.
The agent will use the Datadog MCP connection to pull the relevant data and incorporate it into its analysis.

​
Get Started

👉 AI Player overview 👉 Remote MCP Servers overview 👉 Setup guide