The list of metrics (based on sFlow counter exports) that sFlow-RT can report is listed here:
The actual metrics you will see depends on the capabilities of the sFlow agents sending telemetry to sFlow-RT. The /prometheus/metrics/ALL/ALL/txt REST API call dumps the metrics in Prometheus scrape format using the sflow_ prefix for each metric name. The following example shows how to create a Prometheus/Grafana dashboard of interface counters:
Note: You can be selective about the metrics exported, see the REST API documentation for the /prometheus/metrics/*/*/txt query under the sFlow-RT API tab.
There are also some sFlow-RT health statistics available at /prometheus/analyzer/txt and the following shows how to use them in a dashboard:
Finally, if you want to streamline the process of building an sFlow-RT, Prometheus, Grafana monitoring stack, take a look at the Docker Compose example: