audr-adapter-langfuse 0.1.0__tar.gz

audr_adapter_langfuse-0.1.0/.gitignore

@@ -0,0 +1,7 @@
+dist/
+.venv/
+__pycache__/
+.pytest_cache/
+.ruff_cache/
+.mypy_cache/
+.coverage

audr_adapter_langfuse-0.1.0/CHANGELOG.md

@@ -0,0 +1,16 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]
+
+## [0.1.0] - 2026-11-13
+
+### Added
+
+- Initial release: an OpenTelemetry span processor that turns Langfuse generation,
+  embedding, tool and retriever observations into AUDR records for an existing
+  `audr.Client`.

audr_adapter_langfuse-0.1.0/LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 Chargebee, Inc. and the AUDR contributors
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

audr_adapter_langfuse-0.1.0/NOTICE

@@ -0,0 +1,16 @@
+Agent Usage Detail Record (AUDR)
+Copyright 2026 Chargebee, Inc. and the AUDR contributors
+
+This product includes software developed at Chargebee, Inc.
+(https://www.chargebee.com/).
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use
+these files except in compliance with the License. You may obtain a copy of the
+License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed
+under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR
+CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.

audr_adapter_langfuse-0.1.0/PKG-INFO

@@ -0,0 +1,195 @@
+Metadata-Version: 2.5
+Name: audr-adapter-langfuse
+Version: 0.1.0
+Summary: Langfuse adapter for AUDR (experimental)
+Project-URL: Homepage, https://openaudr.dev
+Project-URL: Issues, https://github.com/openaudr/audr/issues
+Project-URL: Source, https://github.com/openaudr/audr/tree/main/adapters/langfuse/python
+Project-URL: Changelog, https://github.com/openaudr/audr/blob/main/adapters/langfuse/python/CHANGELOG.md
+Project-URL: Specification, https://openaudr.dev/spec/v1.1.0/
+Author-email: The AUDR contributors <audr@chargebee.com>
+License-Expression: Apache-2.0
+License-File: LICENSE
+License-File: NOTICE
+Keywords: adapter,audr,langfuse,llm,metering,opentelemetry
+Classifier: Development Status :: 3 - Alpha
+Classifier: Framework :: AsyncIO
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Typing :: Typed
+Requires-Python: >=3.11
+Requires-Dist: audr<0.3,>=0.2
+Requires-Dist: pydantic<3,>=2.10
+Provides-Extra: runtime
+Requires-Dist: langfuse<6,>=5.2; extra == 'runtime'
+Requires-Dist: opentelemetry-sdk<2,>=1.45; extra == 'runtime'
+Description-Content-Type: text/markdown
+
+# audr-adapter-langfuse
+
+[![PyPI](https://img.shields.io/pypi/v/audr-adapter-langfuse?include_prereleases)](https://pypi.org/project/audr-adapter-langfuse/)
+[![Python versions](https://img.shields.io/pypi/pyversions/audr-adapter-langfuse)](https://pypi.org/project/audr-adapter-langfuse/)
+
+The **AUDR span processor** for [Langfuse](https://langfuse.com) turns every generation,
+embedding, tool and retriever observation recorded through the Langfuse Python SDK into one
+[AUDR](https://openaudr.dev) record for an `audr.Client` your application owns. Because it
+observes the Langfuse layer, one processor meters every integration that reports through
+the SDK: the `langfuse.openai` drop-in, the LangChain `CallbackHandler`,
+`@observe(as_type="generation")` and manual observations. It reads usage, cost,
+identifiers and timings only: never observation input or output, prompts, completions or
+error messages.
+
+> **Status: alpha.** The record model follows AUDR specification 1.x; until this package
+> reaches 1.0.0, a minor release may change the public API.
+
+## Setup
+
+```bash
+pip install "audr-adapter-langfuse[runtime]"
+```
+
+Requires Python 3.11 or later and `langfuse` 5.2 or later within 5.x, which the `runtime`
+extra installs. The Langfuse SDK is built on OpenTelemetry; the adapter is an OpenTelemetry
+`SpanProcessor` added to the same `TracerProvider` as Langfuse's own processor. Langfuse and
+OpenTelemetry are imported when `LangfuseAudrProcessor` is first accessed, not when
+`audr_adapter_langfuse` is imported.
+
+## Usage
+
+Construct the processor on the running event loop that owns the `Client`, add it to a
+`TracerProvider`, and hand that provider to the Langfuse client:
+
+```python
+import asyncio
+
+from audr import Attribution, Client, FileSink
+from audr_adapter_langfuse import LangfuseAudrProcessor, LangfuseConfig
+from langfuse import Langfuse, propagate_attributes
+from opentelemetry.sdk.trace import TracerProvider
+
+
+async def main() -> None:
+    async with Client(FileSink("audr.jsonl")) as client:
+        processor = LangfuseAudrProcessor(
+            client=client,
+            config=LangfuseConfig(
+                provider="openai",
+                attribution_defaults=Attribution(environment="production"),
+            ),
+        )
+        tracer_provider = TracerProvider()
+        tracer_provider.add_span_processor(processor)
+        langfuse = Langfuse(tracer_provider=tracer_provider)  # keys from LANGFUSE_* variables
+        try:
+            with propagate_attributes(metadata={"audr_account_id": "acct_42"}):
+                with langfuse.start_as_current_observation(
+                    name="answer", as_type="generation", model="gpt-4o-mini"
+                ) as generation:
+                    generation.update(usage_details={"input": 120, "output": 40})
+        finally:
+            langfuse.shutdown()
+            await processor.drain(timeout=5)
+            processor.close()
+
+
+asyncio.run(main())
+```
+
+Leaving the `async with` block drains the client's queue and closes its sink. A runnable
+version that keeps Langfuse offline is
+[`examples/observe_generation.py`](https://github.com/openaudr/audr/blob/main/adapters/langfuse/python/examples/observe_generation.py).
+
+> [!IMPORTANT]
+> The processor meters only spans the provider records. Configure the `TracerProvider` you
+> pass to Langfuse to sample every billed trace: Langfuse's own `sample_rate` does not apply
+> to a provider you supply. A Langfuse client that is disabled, because its keys are missing
+> or `tracing_enabled=False`, creates no spans, and no record is produced. Shut down in this
+> order: `langfuse.shutdown()` (or `langfuse.flush()`), `await processor.drain()`,
+> `processor.close()`, then the client; the
+> [shutdown order](https://github.com/openaudr/audr/blob/main/adapters/langfuse/python/docs/reference.md#shutdown-order)
+> explains each step.
+
+## Attribution
+
+Attribution is read for each span from `audr_*` metadata keys, field by field: observation
+metadata wins over that of the observations still open around it, nearest first, which wins
+over trace metadata set with `propagate_attributes()`, which wins over
+`langfuse.environment` (only when it is exactly an AUDR environment), which wins over
+`LangfuseConfig.attribution_defaults`. `labels` merge by key. Langfuse experiment runs
+(`sdk-experiment`) are attributed to `evaluation`. Flat keys keep each value under
+Langfuse's 200-character limit for propagated metadata:
+
+```python
+from langfuse import propagate_attributes
+
+with propagate_attributes(
+    metadata={
+        "audr_environment": "production",
+        "audr_account_id": "acct_42",
+        "audr_subscription_id": "sub_7",
+        "audr_user_id": "u_8f14e45f",  # pseudonymous, never an email or a name
+        "audr_feature_id": "support-chat",  # selects the meter the usage is billed against
+        "audr_labels": {"team": "support"},
+    }
+):
+    pass  # observations created here carry this attribution
+```
+
+- A span with no `environment`, or a `production` span with no `account_id`, is skipped with
+  a warning rather than billed to a guess.
+- Langfuse's own `user_id` is **not** copied to `attribution.user_id`: applications often
+  put an email there. Set `audr_user_id` to a pseudonymous id instead.
+
+## Records
+
+| Langfuse observation type | `resource.operation` | `usage` |
+| --- | --- | --- |
+| `generation` | `generation` | `llm` tokens, `requests: 1` |
+| `embedding` | `embedding` | `llm` tokens, `requests: 1` |
+| `tool` | `tool_execution` | `tool`, `type: invocation`, `call_count: 1` |
+| `retriever` | `retrieval` | `tool`, `type: retrieval`, `call_count: 1` |
+
+Other types (`span`, `agent`, `chain`, `event`, `evaluator`, `guardrail`) and spans from
+other OpenTelemetry instrumentation are not metered.
+
+- **Identifiers.** `run.run_id` and `run.trace_id` are the OpenTelemetry trace id, and
+  `run.span_id` and `run.parent_span_id` the span ids, so every observation in one Langfuse
+  trace joins.
+- **Usage.** Cache reads, cache writes and reasoning are counted apart from
+  `input_tokens` and `output_tokens`. Langfuse's exclusive buckets are folded back together,
+  so audio and modality counts stay in input and output; raw OpenAI usage, as the
+  `langfuse.openai` drop-in records it, and raw Anthropic usage are recognised too. The
+  [reference](https://github.com/openaudr/audr/blob/main/adapters/langfuse/python/docs/reference.md#tokens)
+  lists every key.
+- **Cost.** A generation's or embedding's `cost_details` becomes `cost.total_cost` in USD
+  when the application or integration set it on the span. Langfuse's server-side inferred
+  cost is never seen, so most records carry no cost.
+- **Provider.** Langfuse records no provider. The adapter takes `gen_ai.provider.name`,
+  then `gen_ai.system`, then the LangChain `ls_provider` metadata, then
+  `LangfuseConfig.provider`. A generation or embedding with none is skipped with a warning.
+- **Failures.** A span with status `ERROR` or level `ERROR` carries
+  `run.error_code = LANGFUSE_OBSERVATION_ERROR`. Failed tool and retriever spans are always
+  recorded; a failed generation or embedding is recorded only when it reported usage or a
+  positive cost.
+
+When one model call is recorded as a model span inside another, as with LangChain's
+`ChatOpenAI` while the `langfuse.openai` drop-in is imported, only the inner span is
+recorded, with the `audr_*` metadata of the spans around it. Running this adapter alongside
+another AUDR adapter that observes the same calls, such as the LiteLLM adapter, records each
+call twice.
+
+## Documentation
+
+- [Reference](https://github.com/openaudr/audr/blob/main/adapters/langfuse/python/docs/reference.md): options, span metadata, record fields, shutdown order, diagnostics, operational bounds
+- [Examples](https://github.com/openaudr/audr/tree/main/adapters/langfuse/python/examples): runnable offline, without Langfuse keys
+- [Changelog](https://github.com/openaudr/audr/blob/main/adapters/langfuse/python/CHANGELOG.md)
+- [AUDR specification](https://openaudr.dev), which defines every record field
+
+## License
+
+Apache-2.0. Contributions follow [`CONTRIBUTING.md`](https://github.com/openaudr/audr/blob/main/CONTRIBUTING.md).

audr_adapter_langfuse-0.1.0/README.md

@@ -0,0 +1,163 @@
+# audr-adapter-langfuse
+
+[![PyPI](https://img.shields.io/pypi/v/audr-adapter-langfuse?include_prereleases)](https://pypi.org/project/audr-adapter-langfuse/)
+[![Python versions](https://img.shields.io/pypi/pyversions/audr-adapter-langfuse)](https://pypi.org/project/audr-adapter-langfuse/)
+
+The **AUDR span processor** for [Langfuse](https://langfuse.com) turns every generation,
+embedding, tool and retriever observation recorded through the Langfuse Python SDK into one
+[AUDR](https://openaudr.dev) record for an `audr.Client` your application owns. Because it
+observes the Langfuse layer, one processor meters every integration that reports through
+the SDK: the `langfuse.openai` drop-in, the LangChain `CallbackHandler`,
+`@observe(as_type="generation")` and manual observations. It reads usage, cost,
+identifiers and timings only: never observation input or output, prompts, completions or
+error messages.
+
+> **Status: alpha.** The record model follows AUDR specification 1.x; until this package
+> reaches 1.0.0, a minor release may change the public API.
+
+## Setup
+
+```bash
+pip install "audr-adapter-langfuse[runtime]"
+```
+
+Requires Python 3.11 or later and `langfuse` 5.2 or later within 5.x, which the `runtime`
+extra installs. The Langfuse SDK is built on OpenTelemetry; the adapter is an OpenTelemetry
+`SpanProcessor` added to the same `TracerProvider` as Langfuse's own processor. Langfuse and
+OpenTelemetry are imported when `LangfuseAudrProcessor` is first accessed, not when
+`audr_adapter_langfuse` is imported.
+
+## Usage
+
+Construct the processor on the running event loop that owns the `Client`, add it to a
+`TracerProvider`, and hand that provider to the Langfuse client:
+
+```python
+import asyncio
+
+from audr import Attribution, Client, FileSink
+from audr_adapter_langfuse import LangfuseAudrProcessor, LangfuseConfig
+from langfuse import Langfuse, propagate_attributes
+from opentelemetry.sdk.trace import TracerProvider
+
+
+async def main() -> None:
+    async with Client(FileSink("audr.jsonl")) as client:
+        processor = LangfuseAudrProcessor(
+            client=client,
+            config=LangfuseConfig(
+                provider="openai",
+                attribution_defaults=Attribution(environment="production"),
+            ),
+        )
+        tracer_provider = TracerProvider()
+        tracer_provider.add_span_processor(processor)
+        langfuse = Langfuse(tracer_provider=tracer_provider)  # keys from LANGFUSE_* variables
+        try:
+            with propagate_attributes(metadata={"audr_account_id": "acct_42"}):
+                with langfuse.start_as_current_observation(
+                    name="answer", as_type="generation", model="gpt-4o-mini"
+                ) as generation:
+                    generation.update(usage_details={"input": 120, "output": 40})
+        finally:
+            langfuse.shutdown()
+            await processor.drain(timeout=5)
+            processor.close()
+
+
+asyncio.run(main())
+```
+
+Leaving the `async with` block drains the client's queue and closes its sink. A runnable
+version that keeps Langfuse offline is
+[`examples/observe_generation.py`](https://github.com/openaudr/audr/blob/main/adapters/langfuse/python/examples/observe_generation.py).
+
+> [!IMPORTANT]
+> The processor meters only spans the provider records. Configure the `TracerProvider` you
+> pass to Langfuse to sample every billed trace: Langfuse's own `sample_rate` does not apply
+> to a provider you supply. A Langfuse client that is disabled, because its keys are missing
+> or `tracing_enabled=False`, creates no spans, and no record is produced. Shut down in this
+> order: `langfuse.shutdown()` (or `langfuse.flush()`), `await processor.drain()`,
+> `processor.close()`, then the client; the
+> [shutdown order](https://github.com/openaudr/audr/blob/main/adapters/langfuse/python/docs/reference.md#shutdown-order)
+> explains each step.
+
+## Attribution
+
+Attribution is read for each span from `audr_*` metadata keys, field by field: observation
+metadata wins over that of the observations still open around it, nearest first, which wins
+over trace metadata set with `propagate_attributes()`, which wins over
+`langfuse.environment` (only when it is exactly an AUDR environment), which wins over
+`LangfuseConfig.attribution_defaults`. `labels` merge by key. Langfuse experiment runs
+(`sdk-experiment`) are attributed to `evaluation`. Flat keys keep each value under
+Langfuse's 200-character limit for propagated metadata:
+
+```python
+from langfuse import propagate_attributes
+
+with propagate_attributes(
+    metadata={
+        "audr_environment": "production",
+        "audr_account_id": "acct_42",
+        "audr_subscription_id": "sub_7",
+        "audr_user_id": "u_8f14e45f",  # pseudonymous, never an email or a name
+        "audr_feature_id": "support-chat",  # selects the meter the usage is billed against
+        "audr_labels": {"team": "support"},
+    }
+):
+    pass  # observations created here carry this attribution
+```
+
+- A span with no `environment`, or a `production` span with no `account_id`, is skipped with
+  a warning rather than billed to a guess.
+- Langfuse's own `user_id` is **not** copied to `attribution.user_id`: applications often
+  put an email there. Set `audr_user_id` to a pseudonymous id instead.
+
+## Records
+
+| Langfuse observation type | `resource.operation` | `usage` |
+| --- | --- | --- |
+| `generation` | `generation` | `llm` tokens, `requests: 1` |
+| `embedding` | `embedding` | `llm` tokens, `requests: 1` |
+| `tool` | `tool_execution` | `tool`, `type: invocation`, `call_count: 1` |
+| `retriever` | `retrieval` | `tool`, `type: retrieval`, `call_count: 1` |
+
+Other types (`span`, `agent`, `chain`, `event`, `evaluator`, `guardrail`) and spans from
+other OpenTelemetry instrumentation are not metered.
+
+- **Identifiers.** `run.run_id` and `run.trace_id` are the OpenTelemetry trace id, and
+  `run.span_id` and `run.parent_span_id` the span ids, so every observation in one Langfuse
+  trace joins.
+- **Usage.** Cache reads, cache writes and reasoning are counted apart from
+  `input_tokens` and `output_tokens`. Langfuse's exclusive buckets are folded back together,
+  so audio and modality counts stay in input and output; raw OpenAI usage, as the
+  `langfuse.openai` drop-in records it, and raw Anthropic usage are recognised too. The
+  [reference](https://github.com/openaudr/audr/blob/main/adapters/langfuse/python/docs/reference.md#tokens)
+  lists every key.
+- **Cost.** A generation's or embedding's `cost_details` becomes `cost.total_cost` in USD
+  when the application or integration set it on the span. Langfuse's server-side inferred
+  cost is never seen, so most records carry no cost.
+- **Provider.** Langfuse records no provider. The adapter takes `gen_ai.provider.name`,
+  then `gen_ai.system`, then the LangChain `ls_provider` metadata, then
+  `LangfuseConfig.provider`. A generation or embedding with none is skipped with a warning.
+- **Failures.** A span with status `ERROR` or level `ERROR` carries
+  `run.error_code = LANGFUSE_OBSERVATION_ERROR`. Failed tool and retriever spans are always
+  recorded; a failed generation or embedding is recorded only when it reported usage or a
+  positive cost.
+
+When one model call is recorded as a model span inside another, as with LangChain's
+`ChatOpenAI` while the `langfuse.openai` drop-in is imported, only the inner span is
+recorded, with the `audr_*` metadata of the spans around it. Running this adapter alongside
+another AUDR adapter that observes the same calls, such as the LiteLLM adapter, records each
+call twice.
+
+## Documentation
+
+- [Reference](https://github.com/openaudr/audr/blob/main/adapters/langfuse/python/docs/reference.md): options, span metadata, record fields, shutdown order, diagnostics, operational bounds
+- [Examples](https://github.com/openaudr/audr/tree/main/adapters/langfuse/python/examples): runnable offline, without Langfuse keys
+- [Changelog](https://github.com/openaudr/audr/blob/main/adapters/langfuse/python/CHANGELOG.md)
+- [AUDR specification](https://openaudr.dev), which defines every record field
+
+## License
+
+Apache-2.0. Contributions follow [`CONTRIBUTING.md`](https://github.com/openaudr/audr/blob/main/CONTRIBUTING.md).