@testrelic/maestro-analytics 1.3.2 → 1.4.0-next.121

package/README.md

@@ -121,6 +121,19 @@ console.log(`AI defects: ${result.aiDefects.length}`);
 | `metadata` | `object` | — | Custom key/value pairs attached to the run (e.g. build ID, branch name) |
 | `reportMode` | `'batch'` | `'batch'` | Upload strategy — `'batch'` sends all results at the end of the run |
 
+### What the report shows
+
+The run overview opens with a **verdict band** — what broke, in a sentence, with a proportion bar — followed by tiles for what this run actually captured and a **needs-attention** block that leads with the failing step and the driver-log line recorded beside it.
+
+Four surfaces read data the parser has always collected:
+
+- **Device strip.** `deviceId` and platform reached the cloud payload only; both reports now open with the device, and flow rows carry a device badge. The platform is a field of its own, so it is no longer appended to your tag list.
+- **Step kinds.** All ten Maestro step categories are badged separately rather than flattened to four — including the `AI` and `DEV` badges the renderer already had but could never reach.
+- **Driver log.** `maestro.log` was uploaded to the cloud and never rendered locally. It now has a **Driver log** tab.
+- **Replay.** The screen recording with a scrubber marked once per step; clicking a step seeks the video. `videoOffset` was computed in seconds for cloud video sync and the local report never read it.
+
+Screenshots and AI defects are now scoped to the flow they belong to, rather than showing the whole run on every flow.
+
 ---
 
 ## Cloud Integration