Give an AI agent your Datadog as a tool and it can act on your telemetry instead of only reading it. It can create monitors and SLOs, open an incident, schedule downtime around a risky change, aggregate logs to trace a spike, and tag hosts, calling each operation as a tool while it works a problem end to end. Wire it into a coding agent in NoClick and it can go from a noisy alert to an opened incident and a scoped downtime on its own.
Pick the operations you want and each becomes a tool the agent can call directly while it works.
Create an alert/metric monitor.
Fetch a single monitor's details.
List all monitors, optionally filtered.
Edit an existing monitor.
Delete a monitor.
Silence a monitor, optionally scoped and timed.
Re-enable a muted monitor.
Search monitors by query string.
Post an event to the event stream.
Fetch a single event.
List events in a time range.
Query events with a structured filter.
Submit metric data points (time series).
Cross-product timeseries data query (v2).
Query metric points over a time range (legacy v1).
List metrics actively reporting since a timestamp.
Fetch metadata for a metric.
Submit a log entry to the intake.
Search log events (v2 query).
Aggregate/group log events.
Create a dashboard.
Fetch a dashboard definition.
List all dashboards.
Update a dashboard.
Delete a dashboard.
Open an incident.
List incidents.
Update incident fields/state.
Schedule a monitor downtime.
Cancel a scheduled downtime.
List scheduled downtimes.
Get a downtime by ID.
Update an existing downtime.
Create a Service Level Objective.
Get an SLO by ID.
List all SLOs.
Update an existing SLO.
Delete an SLO.
Get SLO performance history.
List infrastructure hosts.
Get total active host counts.
Mute a host to silence its alerts.
Unmute a host.
Get tags for a specific host.
Add tags to a host.
Replace all tags on a host.
Remove all tags from a host.
Post a service check run.
List synthetic tests.
Get a synthetic test configuration.
Trigger one or more synthetic tests to run immediately.
Pause or resume a synthetic test.
OpenCode is an open, model-agnostic command-line coding agent. In NoClick it runs hosted and gains your apps as tools, so it can act on real systems rather than just edit files. Wire an integration into the agent and OpenCode can use that app’s operations directly while it works. It is a flexible choice when you want an open agent with broad model support and real tool access.
NoClick runs OpenCode for you with Datadog wired in as tools. Connect your account and run.