Pull request overview

Improves Python OpenTelemetry chat span telemetry by backfilling gen_ai.request.model (and the span name) when the request model is initially unknown but the response includes a resolved model, addressing issue #5088’s “unknown request model” gap.

Changes:

• Add ChatTelemetryLayer._backfill_request_model(...) to update REQUEST_MODEL and span name from RESPONSE_MODEL when the request model is "unknown".
• Invoke the backfill step for both streaming and non-streaming chat responses before capturing response attributes.
• Add unit tests covering the helper behavior and end-to-end chat client telemetry updates (streaming + non-streaming).

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

            from agent_framework import Agent
            from agent_framework.foundry import FoundryAgent
            from azure.identity import AzureCliCredential

            client = FoundryAgent(
                project_endpoint="https://your-project.services.ai.azure.com",
                agent_name="my-prompt-agent",
                agent_version="1",
                credential=AzureCliCredential(),
            )

            agent = Agent(client=client, tools=[my_function_tool])

Automated Code Review

Reviewers: 3 | Confidence: 91% | Result: All clear

Reviewed: Correctness, Test Coverage, Design Approach

Automated review by TaoChenOSU's agents

Automated Code Review

Reviewers: 4 | Confidence: 92% | Result: All clear

Reviewed: Correctness, Security Reliability, Test Coverage, Design Approach

Automated review by TaoChenOSU's agents