How to use Datadog with VS Code
You want Datadog inside VS Code so it can investigate a production error, trace a slow request, compare incident volume over time, and correlate logs with a failing trace — without you copy-pasting between tabs. Here is the whole setup, once.
Create your Datadog gateway
The gateway is the thing VS Code connects to, and it is where you decide which Datadog tools an agent may use. You create it once and every client reuses it.
-
Create an Arcade account
Sign up and open the dashboard. The free tier is enough to follow this through.
-
Create an MCP Gateway and select your Datadog tools
In the dashboard, create a gateway and pick the Datadog tools you want to expose. The gateway is what your client connects to, and it is where tool access is decided.
-
Copy your gateway URL
Every client below needs it.
https://api.arcade.dev/mcp/<YOUR-GATEWAY-SLUG>
Connect the gateway to VS Code
-
Add the server
VS Code 1.107.0 or higher. Command Palette → MCP: Add Server… → choose HTTP, paste your gateway URL, and name it something like mcp-arcade.
-
Start it and authenticate
Click Start next to the server in mcp.json or under Extensions → MCP Servers - Installed, then approve the Arcade sign-in and consent screens.
-
Try it out
Open the chat pane, switch to Agent mode, and ask it to use one of your Datadog tools.
Full VS Code guide in the docs steps verified 2026-08-13
Get started with Datadog prompts
VS Code can now act on your Datadog account. Click a prompt to copy it.
Keep the shared API key or secret in Arcade and allow only the read-only searches your agent needs. Pair it with Contextual Access policies to control access before an agent can inspect logs, spans, or traces.