checkly · tnolet · Jun 1, 2026 · Jun 1, 2026 · Jun 1, 2026
diff --git a/ai/rocky-ai.mdx b/ai/rocky-ai.mdx
@@ -4,7 +4,7 @@ description: "Learn about Rocky AI — Checkly's AI agent — and configuring an
 sidebarTitle: "Rocky AI"
 ---
 
-Rocky AI is Checkly's AI agent. Rocky AI can help you automatically determine user impact and root cause on any failing check using error messages, code, check descriptions, metrics, traces and logs.
+Rocky AI is Checkly's AI agent. Rocky AI can help you automatically determine user impact and root cause on any failing check using error messages, code, metrics, OTeL traces, trace routes, packet captures: essentially all artefact generated by a check or test.
 
 ## Rocky AI Capabilities
 

diff --git a/images/rocky_ai_otel_spans.png b/images/rocky_ai_otel_spans.png
diff --git a/resolve/ai-root-cause-analysis/overview.mdx b/resolve/ai-root-cause-analysis/overview.mdx
@@ -21,14 +21,14 @@ Rocky AI — Checkly's Agent — can automatically analyze, categorize and summa
   />
 </Frame>
 
-Using any available artifacts like error message, user code, check description, logs, metrics, trace routes, packet captures, Playwright traces, screenshots and other evidence, it will quickly assess what the impact is of a failure and what the underlying (possible) root cause is. It will also provide a suggested code fix for any check based on (Playwright) code.
+Using any available artifacts like error message, user code, check description, logs, metrics, trace routes, Open Telemetry traces, packet captures, Playwright traces, screenshots and other evidence, it will quickly assess what the impact is of a failure and what the underlying (possible) root cause is. It will also provide a suggested code fix for any check based on (Playwright) code.
 
 ## Getting Started with Rocky AI Root Cause Analysis
 
 To use Rocky AI Root Cause Analysis, make sure [Rocky AI is enabled on its settings page](https://app.checklyhq.com/accounts/rocky-ai).
 
 <Info>
-   Rocky AI Root Cause Analysis is available for all synthetic and monitor check types in both testing and monitoring use cases.
+  Rocky AI Root Cause Analysis is available for all synthetic and monitor check types in both testing and monitoring use cases.
 </Info>
 
 ### Automatic analysis

diff --git a/resolve/traces/overview.mdx b/resolve/traces/overview.mdx
@@ -1,11 +1,13 @@
 ---
-title: Traces
-description: 'Trace and debug requests and data flowing through your application and identify performance bottlenecks and failures.'
-sidebarTitle: 'Overview'
+title: "Traces"
+description: "Trace and debug requests and data flowing through your application and identify performance bottlenecks and failures."
+sidebarTitle: "Overview"
 ---
 
-![Check result with OpenTelemetry trace](/images/docs/images/otel/otel_check_result.png)
-
+<img
+  src="/images/docs/images/otel/otel_check_result.png"
+  alt="Check result with OpenTelemetry trace"
+/>
 
 ## What is Distributed Tracing?
 
@@ -26,51 +28,58 @@ With telemetry traces configured, you will have access to traces in all the plac
 
 To get started with Checkly Traces using OpenTelemetry, pick the scenario that best fits your needs.
 
-<Columns cols={3} >
-<Card title="I don't have an OpenTelemetry setup" href="/resolve/traces/instrumentation/nodejs">
-Instrument your app and send traces directly to Checkly. No need for a 3rd party OTel backend.
-</Card>
-
-<Card title="I already have an OTEL Collector" href="/resolve/traces/import/overview">
-Send your infrastructure traces to Checkly to get contextualized check failure analysis.
-</Card>
+<Columns cols={3}>
+  <Card title="I don't have an OpenTelemetry setup" href="/resolve/traces/instrumentation/nodejs">
+    Instrument your app and send traces directly to Checkly. No need for a 3rd party OTel backend.
+  </Card>
 
-<Card title="I want to export Check results to my OpenTelemetry setup" href="/resolve/traces/export/grafana-tempo">
+  <Card title="I already have an OTEL Collector" href="/resolve/traces/import/overview">
+    Send your infrastructure traces to Checkly to get contextualized check failure analysis.
+  </Card>
 
-Export check results as traces to your 3rd party OTel tooling
-</Card>
+  <Card title="I want to export Check results to my OpenTelemetry setup" href="/resolve/traces/export/grafana-tempo">
+    Export check results as traces to your 3rd party OTel tooling
+  </Card>
 </Columns>
 
-
 See this in action in the video below:
 
 <Frame>
-  <iframe 
-    src="https://www.loom.com/embed/30c143388ba54e9ba6b665dfbfe0d295?sid=8ad8d273-b0bb-48ca-b456-1b137384b9de" 
-    title="Embedded content"
-    className="w-full aspect-video rounded-xl"
-    frameBorder="0"
-    allowfullscreen
-  ></iframe>
+  <iframe src="https://www.loom.com/embed/30c143388ba54e9ba6b665dfbfe0d295?sid=8ad8d273-b0bb-48ca-b456-1b137384b9de" title="Embedded content" className="w-full aspect-video rounded-xl" frameBorder="0" allowfullscreen />
 </Frame>
 
-## Key Benefits
+## Works hand-in-hand with Rocky AI
 
-<AccordionGroup>
-<Accordion title="Cost-Effective Tracing">
-Only traces generated by Checkly checks are forwarded to third-party tools, significantly reducing data transfer costs while maintaining comprehensive monitoring coverage.
-</Accordion>
+Checkly Traces automatically integrates with Rocky AI, helping you automatically analyze root causes of check failures. [Rocky AI Root Cause Analysis](/resolve/ai-root-cause-analysis/overview) can search, fetch and interpret Open Telemetry (OTeL) traces ingested by Checkly Traces and use it as a piece of "evidence" in determining why and where a check failed.\
+\
+A simple example. Your API or Playwright check fails with a 500 error. Luckily your backend is instrumented with Open Telemetry and sends a trace indicating an issue with a downstream database, 3rd party service, proxy etc. Now, Rocky AI automatically pulls these OTeL traces into its context window and correlates it with your failing check. \
+\
+Here's an actual example from Checkly's own backend.
 
-<Accordion title="Unified Observability">
-Combine synthetic monitoring data with application traces in your existing observability platform for a complete picture of system health.
-</Accordion>
+<Frame>
+  ![Rocky Ai Otel Spans](/images/rocky_ai_otel_spans.png)
+</Frame>
 
-<Accordion title="Faster Debugging">
-When checks fail, trace data provides immediate context about what happened in your application stack, accelerating root cause analysis.
-</Accordion>
+1. Our API check that checks if our customer facing Prometheus endpoint works, failed with a 500 error. 
+2. The OTeL trace indicates this was actually do to our **Clickhouse** server returning a 500 errors, immediately telling our on-call team where to start looking.
+3. Without the trace, we would have to look at all the various logging, error tracking and other tools that are integrated into the various middleware and infrastructure this request passes through (our load balancer, REST API server, Redis datastore etc.
 
-<Accordion title="Seamless Integration">
-Works with existing OpenTelemetry instrumentation and popular observability platforms without requiring changes to your application logic.
-</Accordion>
-</AccordionGroup>
+## Key Benefits
 
+<AccordionGroup>
+  <Accordion title="Cost-Effective Tracing">
+    Only traces generated by Checkly checks are forwarded to third-party tools, significantly reducing data transfer costs while maintaining comprehensive monitoring coverage.
+  </Accordion>
+
+  <Accordion title="Unified Observability">
+    Combine synthetic monitoring data with application traces in your existing observability platform for a complete picture of system health.
+  </Accordion>
+
+  <Accordion title="Faster Debugging">
+    When checks fail, trace data provides immediate context about what happened in your application stack, accelerating root cause analysis.
+  </Accordion>
+
+  <Accordion title="Seamless Integration">
+    Works with existing OpenTelemetry instrumentation and popular observability platforms without requiring changes to your application logic.
+  </Accordion>
+</AccordionGroup>