@arizeai/phoenix-mcp

--- description: Create flows using Microsoft PromptFlow and send their traces to Phoenix --- # Prompt Flow Tracing This integration will allow you to trace [Microsoft PromptFlow](https://github.com/microsoft/promptflow) flows and send their traces into[`arize-phoenix`](https://github.com/Arize-ai/phoenix). ## Launch Phoenix {% include "../../../../phoenix-integrations/.gitbook/includes/sign-up-for-phoenix-sign-up....md" %} ## Install ```bash pip install promptflow ``` ## Setup Set up the OpenTelemetry endpoint to point to Phoenix and use Prompt flow's `setup_exporter_from_environ` to start tracing any further flows and LLM calls. ```python import os from opentelemetry.sdk.environment_variables import OTEL_EXPORTER_OTLP_ENDPOINT from promptflow.tracing._start_trace import setup_exporter_from_environ endpoint = f"{os.environ["PHOENIX_COLLECTOR_ENDPOINT]}/v1/traces" # replace with your Phoenix endpoint if self-hosting os.environ[OTEL_EXPORTER_OTLP_ENDPOINT] = endpoint setup_exporter_from_environ() ``` ## Run PromptFlow Proceed with creating Prompt flow flows as usual. See this [example notebook](https://github.com/Arize-ai/openinference/blob/main/python/instrumentation/openinference-instrumentation-promptflow/examples/chat_flow_example_to_phoenix.ipynb) for inspiration. ## Observe You should see the spans render in Phoenix as shown in the below screenshots. ## Resources * [Example Notebook](https://github.com/Arize-ai/openinference/blob/main/python/instrumentation/openinference-instrumentation-promptflow/examples/chat_flow_example_to_phoenix.ipynb)