OTel GenAI Traces for Agent and Tool#6653
Conversation
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #6653 +/- ##
==========================================
- Coverage 79.78% 79.65% -0.13%
==========================================
Files 228 229 +1
Lines 17054 17126 +72
==========================================
+ Hits 13606 13642 +36
- Misses 3448 3484 +36
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
|
@EItanya please take a look |
EItanya
left a comment
There was a problem hiding this comment.
These changes make sense to me overall, I am going to do dig deeper into the other instrumentation libraries now to make sure it makes sense in that context as well
EItanya
left a comment
There was a problem hiding this comment.
My one major comment is about error.type attribute. That seems to be the only stable attribute in these spans. I'm not a Python expert, but can we capture the exception, add it, and then bubble it up?
Yes, this is done. I added try-catch block to set the attribute if an error occurs. |
EItanya
left a comment
There was a problem hiding this comment.
These changes look great, I am going to test them out in kagent today :)
Add OTel GenAI traces:
create_agentinvoke_agnetexecute_toolIntroduces context manager helpers to create these traces. The helpers also serve as instrumentation points for other instrumentation libraries.
Resolves #6644