-
Notifications
You must be signed in to change notification settings - Fork 4
Transcripts #498
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Transcripts #498
Conversation
❌ Deploy Preview for rp-cloud failed. Why did it fail? →
|
|
Important Review skippedDraft detected. Please check the settings in the CodeRabbit UI or the You can disable this status message by setting the
✨ Finishing touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
|
||
| Use the time range selector to focus on specific time periods (from the last five minutes up to the last 24 hours): | ||
|
|
||
| * View recent executions, for example, over the last hour, to monitor real-time activity |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm not sure that transcripts update in real time, is there a better use case for viewing the most recent executions?
|
|
||
| == Inspect span details | ||
|
|
||
| The transcript table displays the following: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I happened to spot "Viewing linked trace / View surrounding traces" in the demo cluster, what triggers that and how is it used?
|
|
||
| == What are transcripts | ||
|
|
||
| Every agent and MCP server automatically emits OpenTelemetry traces to a glossterm:topic[] called `redpanda.otel_traces`. These traces provide detailed observability into operations, creating complete transcripts. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Gateway and knowledgebase as well
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We can't include knowledge bases for now as it is not part of Package 1, but will add when it's scheduled.
| @@ -0,0 +1,355 @@ | |||
| = Transcripts and AI Observability | |||
| :description: Understand how Redpanda captures execution transcripts for agents and MCP servers using OpenTelemetry. | |||
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Gateways, Knowledgebases and all future components will use as well.
| :learning-objective-2: Interpret transcript structure for debugging and monitoring | ||
| :learning-objective-3: Distinguish between transcripts and audit logs | ||
|
|
||
| Redpanda automatically captures transcripts (also referred to as execution logs or traces) for both AI agents and MCP servers, providing complete observability into how your agentic systems operate. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
FWIW I don't see the value of calling it transcripts since we have to explain otel traces here in this docs. Why not stick to the actual technical name here since we have to bother our users with traces, spans and the otel convs (when it comes to bring your own agents). IMO introduces unncessary confusion. cc @splittingfield
- Fix endpoint URL format: use pipeline URL pattern instead of internal ports (4317/4318) - Simplify pipeline YAML config with Cloud-standard env vars - Add authentication section with link to Cloud API auth docs - Add Go examples for both HTTP and gRPC transports - Add gRPC examples for Python and Node.js - Remove internal implementation details users don't need Co-authored-by: Johannes Brüderl <johannes@redpanda.fritz.box>
Description
This pull request adds comprehensive documentation for agent and MCP server execution transcripts, enhancing observability in Redpanda using OpenTelemetry traces. It introduces a new "Transcripts" section to the documentation, including a conceptual overview, details on transcript structure and hierarchy, and guidance on how traces are stored and used for debugging and monitoring.
modules/ROOT/nav.adoc, including conceptual, usage, and integration topics for agent and MCP server execution transcripts.modules/ai-agents/pages/observability/index.adocas an entry point for transcript documentation, summarizing the purpose and scope of OpenTelemetry-based monitoring.modules/ai-agents/pages/observability/concepts.adoc, providing an in-depth explanation of how Redpanda captures and organizes execution transcripts, including span types, hierarchy, instrumentation layers, and key attributes for both agents and MCP servers. It also clarifies the distinction between transcripts and audit logs and describes how trace data is stored and managed.modules/ai-agents/pages/observability/transcripts.adocto guide users on navigating the Transcripts UIResolves https://github.com/redpanda-data/documentation-private/issues/
Review deadline: 5 Feb
Page previews
Checks