fix(telemetry): export cache token metrics as strings #1000
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Convert all gen_ai.usage token metrics to strings to ensure
consistent export format in OpenTelemetry traces. Previously,
when cache_write_input_tokens or cache_read_input_tokens were
zero, they were exported as objects {intValue: 0} instead of
string "0", causing issues in trace analysis tools like Langfuse.
Changes:
- Convert all token metrics to strings in end_model_invoke_span
- Convert all token metrics to strings in end_agent_span
- Update test assertions to expect string values
Fixes strands-agents#971
ascender1729
requested a deployment
to
manual-approval
GitHub Actions
Waiting
poshinchen
reviewed
Oct 9, 2025
There was a problem hiding this comment.
https://opentelemetry.io/docs/specs/semconv/registry/attributes/gen-ai/#gen-ai-input-messages
These values are meant to be int though, it is the platform that decides to re-map it as string. We should keep it as int within strands.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Problem
Cache token metrics are exported as objects instead of strings when zero. For example,
gen_ai.usage.cache_write_input_tokensshows as{intValue: 0}instead of"0", causing issues in trace analysis tools like Langfuse.Changes
Convert all token usage metrics to strings in tracer.py:
gen_ai.usage.prompt_tokensgen_ai.usage.completion_tokensgen_ai.usage.cache_read_input_tokensgen_ai.usage.cache_write_input_tokensUpdated test assertions accordingly.
Testing
All 62 tracer tests pass. Verified that zero cache tokens now export as
"0"(string) instead of objects.Fixes #971