Authenticate requests with DD-API-KEY and DD-APPLICATION-KEY headers using YOUR_API_KEY and YOUR_APP_KEY
POST to /api/v1/slo with a JSON body containing type (metric or monitor), name, description, thresholds array (target, warning, timeframe), and the query or monitor_ids
For a metric-based SLO supply numerator and denominator DataDog metric queries in the query object; for a monitor-based SLO supply an array of monitor IDs
GET /api/v1/slo/{slo_id}/history?from_ts=&to_ts= to retrieve historical SLO status and remaining error budget
PATCH /api/v1/slo/{slo_id} to update thresholds or description without recreating the SLO
Use GET /api/v1/slo?ids= to list SLOs and pipe results into a dashboard widget using the SLO widget type in the Datadog dashboard API
Known gotchas
Monitor-based SLOs are tied to the monitor's evaluation window; deleting or modifying the underlying monitor can corrupt the SLO history
The timeframe field in thresholds accepts 7d, 30d, and 90d only; custom rolling windows require a metric-based SLO with a manual query
SLO corrections (planned downtime exclusions) are a separate resource at /api/v1/slo/correction and must be deleted independently of the SLO object
Give your agent this knowledge — and 200+ more routes
One MCP install gives any agent live access to the full route map, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp