Add missing requirements.txt, core module libraries, and bedrock-agents-config.json to support Bedrock agent observability example
This PR addresses https://github.com/langfuse/langfuse-docs/issues/1445, which reports missing files required to reproduce the Bedrock agent integration example from the Langfuse documentation.
- Adds the missing
requirements.txtfile listing necessary dependencies - Includes core module with
instrument_agent_invocationandflush_telemetryfunctions used by the example notebook - Adds
bedrock-agents-config.jsonto support agent and langfuse setup - Updated the example notebook to streamline the flow and make better use of these components
These components were sourced from the AWS-provided sample at: https://github.com/awslabs/amazon-bedrock-agent-samples/tree/main/examples/agent_observability/OpenTelemetry-Agent-Instrumentation
[!IMPORTANT] Adds missing files and modules for Bedrock agent observability, including OpenTelemetry integration and configuration, sourced from AWS samples.
- Files Added:
requirements.txtwith dependencies for Bedrock agent observability.- Core modules in
core/for OpenTelemetry integration, includingagent.py,configuration.py,constants.py,handlers.py,processes.py,streaming_wrapper.py,timer_lib.py,tracing.py, andutils.py.bedrock-agents-config.jsonfor agent and Langfuse setup.- Functionality:
- Implements OpenTelemetry tracing for Bedrock agent using
SpanManagerinagent.py.- Provides utility functions and constants for tracing in
tracing.pyandconstants.py.- Handles various trace events in
handlers.pyandprocesses.py.- Supports streaming responses with
streaming_wrapper.py.- Manages timing with
timer_lib.py.- Example Update:
- Updates example notebook to integrate new components for improved flow.
- Source:
- Components sourced from AWS sample repository for Bedrock agent.
This description was created by
for 3da36393f9640e84c2b520b707c7548c9a72ac32. It will automatically update as commits are pushed.
@aboavent is attempting to deploy a commit to the langfuse Team on Vercel.
A member of the Team first needs to authorize it.
Check out this pull request on ![]()
See visual diffs & provide feedback on Jupyter Notebooks.
Powered by ReviewNB
Heads-up: Vercel deployment requires manual approval from the langfuse team since this PR comes from a fork. Could someone from the team please approve the deployment?
Thanks a lot for addressing the issue