Docs

Connect Datadog.

Connect Datadog metrics and log aggregates by authorizing Perfloop in Datadog. To include Go CPU profiles, use API and application keys.

Access you need

Setup

1 · choose hosted or manual access

In Perfloop Setup, select Datadog. Use provider authorization for metrics and logs. Select manual setup when Go CPU profiles are required.

2 · authorize access

Choose your Datadog site. Select Connect with Datadog, complete consent, and return to Setup. For manual setup, select Use API and application keys and enter the keys.

3 · select scopes and save

Add each metric, log, or profile tag scope that Perfloop may read, then select Save connection. Perfloop checks the selected resources before it saves the connection.

An Admin's coding agent can return a Setup link that opens this provider's form. Complete the form in Setup, and never paste a credential into chat or an MCP call.

Data Perfloop reads

Metric labels and grouping values can contain sensitive data. Review the tags and fields in the resources you connect.

Repo access

Work on any repo in the workspace can query the selected scopes of every telemetry connection in the workspace. Keep telemetry that some repos must not read in a separate workspace.

Credentials and network access

Perfloop encrypts the credential and uses it only for requests within your connection's scope. Models and execution sandboxes do not receive it.

Requests go to the provider's public HTTPS endpoint.

Change or remove access

Authorize Perfloop again or replace the manual connection in Setup to rotate access or change the selected resources. Revoke the OAuth grant or keys in Datadog to stop later reads.

To remove a source, select Delete on its row and confirm. Perfloop deletes the credential it stores, and scheduled checks on the source stop running. Their saved queries and past results stay.

Provider documentation

Security questions: security@perfloop.ai