# Anthropic / Claude

Auto-instrument the Anthropic SDK with the OpenLLMetry instrumentor and send GenAI spans to Coralogix AI Center.

Before you start

These examples export to a local OpenTelemetry Collector over OTLP/gRPC. Deploy the collector first — see [Code examples → Deploy an OpenTelemetry Collector](https://coralogix.com/docs/user-guides/ai/otel-integration/code-examples/#prerequisites-deploy-an-opentelemetry-collector).

## Install

```bash
pip install anthropic opentelemetry-instrumentation-anthropic opentelemetry-sdk opentelemetry-exporter-otlp-proto-grpc
```

## Environment variables

```bash
export ANTHROPIC_API_KEY="<YOUR_ANTHROPIC_API_KEY>"
export OTEL_EXPORTER_OTLP_ENDPOINT="http://<COLLECTOR_HOST>:4317"
export OTEL_EXPORTER_OTLP_INSECURE="true"
export OTEL_RESOURCE_ATTRIBUTES="cx.application.name=my-genai-app,cx.subsystem.name=my-service"
export OTEL_SEMCONV_STABILITY_OPT_IN="gen_ai_latest_experimental"
export TRACELOOP_TRACE_CONTENT="true"
```

## Script

```python
import anthropic

# --- OTel imports ---
from opentelemetry import trace
from opentelemetry.exporter.otlp.proto.grpc.trace_exporter import OTLPSpanExporter
from opentelemetry.instrumentation.anthropic import AnthropicInstrumentor
from opentelemetry.sdk.resources import Resource
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import BatchSpanProcessor


# --- OTel setup: configure tracer provider and OTLP exporter ---
def configure_otel() -> TracerProvider:
    resource = Resource.create()
    provider = TracerProvider(resource=resource)
    exporter = OTLPSpanExporter()  # reads OTEL_EXPORTER_OTLP_ENDPOINT from env
    provider.add_span_processor(BatchSpanProcessor(exporter))
    trace.set_tracer_provider(provider)
    return provider


def main():
    # OTel: initialize tracing and auto-instrument the Anthropic SDK
    provider = configure_otel()
    AnthropicInstrumentor().instrument(tracer_provider=provider)

    # Your app logic
    client = anthropic.Anthropic()
    message = client.messages.create(
        model="claude-sonnet-4-20250514",
        max_tokens=256,
        messages=[
            {"role": "user", "content": "What is OpenTelemetry in one sentence?"}
        ],
    )
    print(f"Claude response: {message.content[0].text}")

    # OTel: flush and shut down the tracer provider
    provider.force_flush()
    provider.shutdown()


if __name__ == "__main__":
    main()
```

## Expected span attributes

- `gen_ai.system` = `"anthropic"`
- `gen_ai.request.model` = `"claude-sonnet-4-20250514"`
- `gen_ai.usage.input_tokens`, `gen_ai.usage.output_tokens`
- `gen_ai.response.finish_reasons` = `["end_turn"]`

Warning

`opentelemetry-instrumentation-anthropic` on PyPI is from OpenLLMetry (Traceloop), not official OTel contrib. Semconv migration from legacy to new is in progress. Content capture uses `TRACELOOP_TRACE_CONTENT=true` (not the standard `OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT`).

## Next steps

Look up which open-source library to use for your provider in [Compatibility matrix](https://coralogix.com/docs/user-guides/ai/otel-integration/providers/index.md).