> ## Documentation Index
> Fetch the complete documentation index at: https://docs.trulayer.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Anthropic

> Trace every Claude API call with one function.

## Install

<CodeGroup>
  ```bash Python theme={null}
  pip install trulayer anthropic
  ```

  ```bash TypeScript theme={null}
  npm install @trulayer/sdk @anthropic-ai/sdk
  ```
</CodeGroup>

## Instrument

<CodeGroup>
  ```python Python theme={null}
  import os
  from anthropic import Anthropic
  import trulayer

  trulayer.init(api_key=os.environ["TRULAYER_API_KEY"], project_name="my-app")

  client = Anthropic()
  trulayer.instrument_anthropic(client)
  ```

  ```typescript TypeScript theme={null}
  import Anthropic from "@anthropic-ai/sdk";
  import { TruLayer, instrumentAnthropic } from "@trulayer/sdk";

  const tl = new TruLayer({
    apiKey: process.env.TRULAYER_API_KEY!,
    projectName: "my-app",
  });

  const anthropic = instrumentAnthropic(new Anthropic(), tl);
  ```
</CodeGroup>

## What gets captured

Every call to `messages.create` (sync and streaming, async and sync) becomes an `llm` span with:

* `input` — the full `messages` array and `system` prompt
* `output` — the response content blocks (text + tool\_use)
* `model`
* `prompt_tokens` → Anthropic's `input_tokens`
* `completion_tokens` → Anthropic's `output_tokens`
* `cache_creation_input_tokens`, `cache_read_input_tokens` → attached as metadata when prompt caching is in use
* `stop_reason` — attached as metadata (`end_turn`, `tool_use`, `max_tokens`)
* `latency_ms`

## Streaming

```python theme={null}
with client.messages.stream(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hi"}],
) as stream:
    for text in stream.text_stream:
        print(text, end="")
# span closes when the `with` block exits with final token counts
```

## Tool use

Claude's tool calls appear in the span output under `content[].tool_use`. To trace tool **execution** as its own span:

```python theme={null}
with trulayer.current_trace().span("tool:get_weather", span_type="tool") as span:
    span.set_input(tool_input)
    result = get_weather(**tool_input)
    span.set_output(result)
```

Then feed the result back to Claude with a follow-up `messages.create` call — that becomes a second `llm` span, and the two-turn conversation appears as a clear waterfall in the dashboard.

## Prompt caching

Anthropic's [prompt caching](https://docs.claude.com/en/docs/build-with-claude/prompt-caching) works transparently — the SDK records `cache_creation_input_tokens` and `cache_read_input_tokens` as span metadata so you can see cache hit rates in the dashboard's Metrics view.

Filter by `metadata.cache_hit_rate` or query via the API:

```bash theme={null}
curl "https://api.trulayer.ai/v1/metrics?project_id=my-app&from=...&filters[model]=claude-sonnet-4-6" \
  -H "Authorization: Bearer $TRULAYER_API_KEY"
```

## Disabling

```python theme={null}
trulayer.uninstrument_anthropic(client)
```

## Known gotchas

* **Extended thinking** — thinking content is captured as part of the span output (`content[].thinking`). It'll show up in trace detail views alongside the visible response.
* **Batch API (`messages.batches`)** — not auto-instrumented. Wrap batch submissions manually with `trace()` if you want observability.
* **Vertex and Bedrock** — use the respective Anthropic SDK client (`AnthropicVertex`, `AnthropicBedrock`) and the same `instrument_anthropic()` call.