2.70.0

New Features ✨

• New integration: Mistral AI (#4733) by @alexander-alderman-webb

  Add the Mistral integration to your sentry_sdk.init call:

import sentry_sdk
from sentry_sdk.integrations.mistral import MistralIntegration

sentry_sdk.init(
    dsn="...",
    traces_sample_rate=1.0,
    integrations=[
        MistralIntegration(),
    ]
)

• New option: A fine-grained way to configure what data should be sent from the SDK.

  Offers much more flexibility than send_default_pii, allowing you to specify what
  data should be collected. If data_collection is defined, it takes precedence over
  send_default_pii.

  See https://docs.sentry.io/platforms/python/configuration/options/#data_collection
  for the available configuration options.

import sentry_sdk

sentry_sdk.init(
    data_collection={
        "user_info": False,
        "gen_ai": {"inputs": False, "outputs": False},
        "graphql": {"document": False, "variables": False},
        "database_query_data": False,
        "queues": False,
        "http_bodies": [],
        "cookies": {
            "mode": "denylist",
            "terms": ["forwarded", "-ip", "remote-", "via", "-user"],
        },
        "http_headers": {
            "request": {
                "mode": "denylist",
                "terms": ["forwarded", "-ip", "remote-", "via", "-user"],
            },
        },
        "url_query_params": {
            "mode": "denylist",
            "terms": ["forwarded", "-ip", "remote-", "via", "-user"],
        },
    }
)

Mistral

• Record gen_ai.output.messages by @alexander-alderman-webb in #7549
• Record gen_ai.input.messages by @alexander-alderman-webb in #7548
• Record gen_ai.system_instructions by @alexander-alderman-webb in #7547
• Record request parameters by @alexander-alderman-webb in #7531
• Record token usage by @alexander-alderman-webb in #7529
• Add integration with Chat.complete and Chat.complete_async patches by @alexander-alderman-webb in #7528

Internal Changes 🔧

• (data-collection) Promote data_collection from _experiments o top-level by @ericapisani in #7560
• Close client on teardown by @alexander-alderman-webb in #7562
• 🤖 Update test matrix with new releases (09/21) by @github-actions in #7546
• Remove getsentry/testing-ai-sdk-integrations by @alexander-alderman-webb in #7526

2.69.2

Bug Fixes 🐛

• (aws-lambda) Attach user info when streaming spans by @sentrivana in #7429
• (bottle) Avoid infinite serializing loop by @sentrivana in #7511
• (clickhouse_driver) Only set db.result in breadcrumb with sensitive data opt-in by @alexander-alderman-webb in #7432
• (google-genai) Tolerate None candidates by @alexander-alderman-webb in #7519
• (pydantic-ai) Stop capturing tool execution spans when validation fails by @alexander-alderman-webb in #7448
• (serializer) Add type annotations to databag limits by @8rulerstar in #7505
• (tracing) Make functions_to_trace work when streaming spans by @sentrivana in #7430
• (typing) Fix disabled_integrations type by @sentrivana in #7506
• Don't attach stacktrace to check-ins by @sentrivana in #7497

Documentation 📚

• Update span streaming docstrings by @sentrivana in #7463

Internal Changes 🔧

Huggingface Hub

• Parametrize tests on the streaming trace lifecycle by @alexander-alderman-webb in #7436
• Only check relevant data with ApproxDict by @alexander-alderman-webb in #7434

Pydantic Ai

• Remove vacuous _set_model_data() tests by @alexander-alderman-webb in #7453
• Remove vacuous _set_agent_data() tests by @alexander-alderman-webb in #7450

Other

• (deps) Lock file maintenance by @renovate in #7302
• (django) Fix flaky cache tests by using uuid for cache LOCATION by @sentry-junior in #7444
• (pymongo) Raise maxWireVersion by @alexander-alderman-webb in #7473
• Remove unused base64 utils by @sentrivana in #7495
• 🤖 Update test matrix with new releases (09/14) by @github-actions in #7486
• 🤖 Update test matrix with new releases (09/07) by @github-actions in #7399

Other

• Skip formatting log records when Sentry logs are off by @gyanu2507 in #7422

2.69.1

New Features ✨

• (django) Add http.route attribute by @alexander-alderman-webb in #7428

Bug Fixes 🐛

• (aiohttp) Preserve SigV4-signed propagation headers by @Robinbinu in #7427

Internal Changes 🔧

Mcp

• Parametrize tests on the streaming trace lifecycle by @alexander-alderman-webb in #7424
• Remove vacuous tests by @alexander-alderman-webb in #7420

Other

• (cohere) Parametrize tests on the streaming trace lifecycle by @alexander-alderman-webb in #7419
• (openai-agents) Parametrize tests on stream_gen_ai_spans and the streaming trace lifecycle by @alexander-alderman-webb in #7405

2.69.0

New Features ✨

Other

• (aiohttp) Add http.route attribute by @alexander-alderman-webb in #7341
• (bottle) Add http.route attribute by @alexander-alderman-webb in #7343
• (clickhouse_driver) Set breadcrumbs in the streaming trace lifecycle by @alexander-alderman-webb in #7325
• (docs) Add Plausible analytics to Sphinx docs by @ericapisani in #7319
• (falcon) Add http.route attribute by @alexander-alderman-webb in #7340
• (fastapi) Add http.route attribute by @alexander-alderman-webb in #7322
• (flask) Add http.route attribute by @alexander-alderman-webb in #7344
• (pyramid) Add http.route attribute by @alexander-alderman-webb in #7347
• (quart) Add http.route attribute by @alexander-alderman-webb in #7348
• (sqlalchemy) Add more db system names for span data by @ericapisani in #7329
• (starlette) Add http.route attribute by @alexander-alderman-webb in #7338
• (types) Export SpanJSON type by @ericapisani in #7331
• Add Python 3.15 to test matrix by @sentrivana in #7326

Bug Fixes 🐛

Mcp

• Handle None arguments in tool and prompt handlers by @alexander-alderman-webb in #7387
• Don't report stateless StreamableHTTP servers as stdio by @Snaylaker in #7224

Openai

• Read input_text parts for the Responses API by @alexander-alderman-webb in #7333
• Check Response.output is a list before iterating by @alexander-alderman-webb in #7238

Starlette

• Use url transaction source for host routing by @mjq in #7266
• Set transaction name on current scope in sync handler by @mjq in #7201

Other

• (batcher) Do not let a failed flush kill the flusher thread (#7138) by @zkasuran in #7186
• (boto3) Fix botocore SigV4 failures caused by post-sign trace propagation by @pabloDeputter in #7050
• (httpx,httpx2) Always capture breadcrumb by @sentrivana in #7351
• (mcp,langchain) Add mechanism to captured exceptions by @gmassello in #7226
• (openai-agents) Use isinstance to determine tool type by @alexander-alderman-webb in #7412
• (quart) Skip quart-flask-patch and adjust error test for quart>=0.23 by @ericapisani in #7320
• Allow to correctly compare parsed versions by @sentrivana in #7184

Internal Changes 🔧

Openai Agents

• Capture spans instead of reading from span recorder by @alexander-alderman-webb in #7404
• Pin openai version when incompatible by @alexander-alderman-webb in #7236

Pydantic Ai

• Replace individual hooks with hooks.on.model_request by @alexander-alderman-webb in #7327
• Fix instructions test for pydantic-ai >=2.36.0 by @ericapisani in #7321

Other

• (aiohttp) Remove dead segment name code by @mjq in #7284
• (anthropic) Include StreamedSpan in type annotations by @alexander-alderman-webb in #7373
• (mcp) Inline _extract_handler_data_from_args() by @alexander-alderman-webb in #7386
• (quart) Fix flake caused by flushing spans on segment end by @alexander-alderman-webb in #7240
• (stdlib) Remove urllib3 workaround by @pabloDeputter in #7283
• Add openai-agents to uv typing group by @alexander-alderman-webb in #7415
• Add aiohttp to uv typing group by @alexander-alderman-webb in #7395
• Isolate httpx test suite from httpx2 by @ZaafirRizwan in #7398
• Remove test_langchain_embeddings_error_handling() by @alexander-alderman-webb in #7389
• Add tiktoken to uv typing group by @alexander-alderman-webb in #7394
• Reference renamed coverage action by @MathurAditya724 in #7371
• Remove flaky detector by @sl0thentr0py in #7363
• Add openai to uv typing group by @alexander-alderman-webb in #7334
• 🤖 Update test matrix with new releases (08/31) by @github-actions in #7303
• Add anthropic to uv typing group by @alexander-alderman-webb in #7324
• Add google-genai to uv typing group by @alexander-alderman-webb in #7323
• Use tagged versions for getsentry/codecov-action by @alexander-alderman-webb in #7241
• Update test matrix with new releases (08/24) by @alexander-alderman-webb in #7228
• Add httpx dependency to tests relying on get_model_response() by @alexander-alderman-webb in #7235
• Access the 'type' in an event using .get by @ericapisani in #7239

2.68.1

enable_logs

• (logs) Don't stop sending auto-collected logs when enable_logs=True by @sentrivana in #7237

  If you have enable_logs set to True, our logging integrations for the standard library logging module as well as Loguru will auto-collect logs and send them to Sentry as Sentry logs by default, preserving old behavior. Turning automatic collection off for a specific integration can still be achieved using the capture_sentry_logs integration option.

  import sentry_sdk
  from sentry_sdk.integrations.logging import LoggingIntegration
  from sentry_sdk.integrations.loguru import LoguruIntegration
  
  sentry_sdk.init(
      enable_logs=True,
      integrations=[
          LoggingIntegration(capture_sentry_logs=True),
          LoguruIntegration(capture_sentry_logs=False),
      ],
  )

  Please note that the enable_logs option is deprecated and will be removed in the next major release. The sentry_sdk.logger.X API now works regardless of it, and auto-collection can be opted into via the capture_sentry_logs integration-level options, which are False by default, unless you have enable_logs=True. We've added this compatibility layer to make the transition to a enable_logs-free world easier.

Bug Fixes 🐛

• (django) Add failed_request_status_codes by @mgaligniana in #7140
• (anthropic) Gate gen_ai.response.tool_calls on outputs, not inputs by @ericapisani in #7207
• (google_genai) Gate streaming gen_ai.response.tool_calls on outputs, not inputs by @ericapisani in #7210
• (langchain) Use gen_ai.tool.definitions attribute when data collection is enabled by @ericapisani in #7204
• (langgraph,huggingface_hub) Gate gen_ai.response.tool_calls on outputs, not inputs by @ericapisani in #7208
• (openai) Gate gen_ai.response.tool_calls on outputs, not inputs by @ericapisani in #7205
• (openai_agents) Gate gen_ai.response.tool_calls on outputs, not inputs by @ericapisani in #7209

2.68.0

Important

• We're making enable_logs and enable_metrics no-op with this release (#7177), and they'll be dropped in the next major.

  Previously, enable_logs also controlled automatic logs collection from the logging and Loguru integrations. These integrations now get an integration-level capture_sentry_logs boolean option to allow for more control over the auto-collection. These options are False by default, i.e., nothing is auto-collected without your explicit opt-in.

  Action Needed

  If you had enable_logs set to True:

  • If you were using the sentry_sdk.logger.X API, no action necessary, the API will just work.
  • If you were auto-collecting logs from either LoggingIntegration or LoguruIntegration, the auto-collection will be turned off in this release. You can switch auto-collection on explicitly with:

  import sentry_sdk
  from sentry_sdk.integrations.logging import LoggingIntegration
  from sentry_sdk.integrations.loguru import LoguruIntegration
  
  sentry_sdk.init(
      integrations=[
          LoggingIntegration(capture_sentry_logs=True),
          LoguruIntegration(capture_sentry_logs=True),
      ],
  )

  If you had enable_logs set to False:

  • If you were using it to gate usages of the sentry_sdk.logger.X API, you'll need to remove the calls entirely or define a before_send_log callback to filter out unwanted logs.

  If you has enable_metrics set to False:

  • Any metrics emitted using the metrics API will be emitted. You'll need to drop them in a before_send_metric or remove the calls to the API.

  Why We're Doing This

  We recognize this is a disruptive change for some folks and want to make it clear this is a one-off. We're removing the options because they were an unnecessary hurdle that one had to jump through to be able to use logs and metrics, and it was confusing why the logging API would not just work on its own. On the other hand, we wanted to give you more fine-grained control over automatic collection.

New Features ✨

Other

• Set the segment source on the scope for NoOpStreamedSpan by @alexander-alderman-webb in #7163
• Remove early return for NoOpStreamedSpan in set_transaction_name() by @alexander-alderman-webb in #7164
• Set the segment name on the scope with NoOpStreamedSpan by @alexander-alderman-webb in #7162

Bug Fixes 🐛

• (batcher) Add global flush trigger based on the span number by @alexander-alderman-webb in #7150

Internal Changes 🔧

HTTPX, HTTPX2

• Move crumbs to integrations by @sentrivana in #7149
• Split into sync/async tests by @sentrivana in #7152

Other

• (aiohttp) Move crumbs to integration by @sentrivana in #7135
• (boto3) Move crumbs to integration by @sentrivana in #7165
• (django) Remove dead code and stale markers from the Django test suite by @ericapisani in #7156
• (pyreqwest) Move crumbs to integration by @sentrivana in #7148
• (stdlib) Move crumbs to integration by @sentrivana in #7161
• Flush trace buckets when segment spans finish by @alexander-alderman-webb in #7170
• Revert global span batcher limits by @alexander-alderman-webb in #7168

2.67.1

Bug Fixes 🐛

• (batcher) Add global flush trigger based on the span number by @alexander-alderman-webb in #7150

2.67.0

New Features ✨

Batcher

• Add global flush trigger based on summed size estimates by @alexander-alderman-webb in #7144
• Add hard span limit by @alexander-alderman-webb in #7143

Integrations

• Add span for DRF authentication by @mgaligniana in #6984

Langchain

• Record reasoning token usage attribute by @alexander-alderman-webb in #7136
• Record cached token usage attributes by @alexander-alderman-webb in #7132

Openai

• Set gen_ai.tool.definitions for the Responses API by @alexander-alderman-webb in #6951
• Set gen_ai.tool.definitions for the Chat Completions API by @alexander-alderman-webb in #6950
• Add gen_ai.request.reasoning.level to OpenAI spans by @pabloDeputter in #6892

Streaming

• Support ai_track decorator by @sentrivana in #6895
• Promote before_send_span to top-level option by @sentrivana in #6885
• Make set_conversation_id API work by @sentrivana in #6866

Other

• (api) Add top-level set_attributes API by @sentrivana in #6897
• (aws-lambda) Enable Python 3.14 support to the Lambda layer by @pabloDeputter in #6916
• (openai-agents) Set the gen_ai.tool.definitions attribute by @alexander-alderman-webb in #6956
• (pydantic-ai) Record gen_ai.output.messages including ThinkingPart by @alexander-alderman-webb in #7037
• Add possibility to allowlist integrations in toxgen by @sentrivana in #6996

Bug Fixes 🐛

Openai Agents

• Patch correct run_single_turn_streamed reference by @alexander-alderman-webb in #7032
• Stop destructively wrapping tools by @alexander-alderman-webb in #6961

Tests

• Adapt to new exception propagation behavior in Quart by @sentrivana in #6995
• Remove redundant type filters after capture_items with args by @sentrivana in #6867

Other

• (aws-lambda) Fix flake in aws lambda tests by @pabloDeputter in #6937
• (langchain) Stop double-counting usage in multi-candidate responses by @alexander-alderman-webb in #7064
• (mcp) Stop adding the prompt name to arguments by @alexander-alderman-webb in #6765
• (pydantic-ai) Stop recording duplicate spans by @alexander-alderman-webb in #7056
• (sampling) sentry-trace should take precedence over baggage by @sentrivana in #6863
• (streaming) Set segment source if unset by @sentrivana in #6791
• Do not let exceptions in before_breadcrumb bubble up by @sentrivana in #7133
• Correct typo in test function name by @TheGreatApollyon in #6967
• Include query string and fragment in url.full span attribute by @sentrivana in #6861

Documentation 📚

• Remove AI commit attribution requirement from AGENTS.md by @ericapisani in #6828

Internal Changes 🔧

Fastmcp

• Skip SSE test for fastmcp>=4 by @alexander-alderman-webb in #6923
• Remove assertions on RootModel instances by @alexander-alderman-webb in #6909
• Replace broad try...except with skipif by @alexander-alderman-webb in #6922

Mcp

• Replace instance patches with middleware by @alexander-alderman-webb in #6764
• Remove redundant mcp.tool.result.is_error by @alexander-alderman-webb in #6915
• Use mcp_types package by @alexander-alderman-webb in #6763
• Inline _set_span_output_data() by @alexander-alderman-webb in #6762
• Inline _get_span_config() by @alexander-alderman-webb in #6761
• Inline _prepare_handler_data() by @alexander-alderman-webb in #6758
• Separate tool, prompt and resource patches by @alexander-alderman-webb in #6726

Openai Agents

• Stop setting gen_ai.response.model on Invoke Agent spans by @alexander-alderman-webb in #7028
• Use first class tool hooks when available by @alexander-alderman-webb in #6994
• Assert subset of expected dictionary is in gen_ai.response.tool_calls by @alexander-alderman-webb in #6973
• Stop setting gen_ai.request.available_tools on Invoke Agent spans by @alexander-alderman-webb in #6955
• Stop setting gen_ai.request.available_tools on Execute Tool spans by @alexander-alderman-webb in #6954

Other

• (ai) Route AI integration tests through OpenRouter by @sentry-junior in #6979
• (arq) Avoid race between the exception and transaction events by @alexander-alderman-webb in #7052
• (aws_lambda) Add ARM64 architecture support for local Lambda testing by @ericapisani in #7053
• (deps) Lock file maintenance by @renovate in #6734
• (django) Use iscoroutinefunction shim in patch_views() by @alexander-alderman-webb in #7024
• (httpx) Replace deprecated asyncio.iscoroutinefunction by @alexander-alderman-webb in #7023
• (huggingface-hub) Fix chat completion stream type by @alexander-alderman-webb in #7070
• (langchain) Remove on_agent_finish() hook by @alexander-alderman-webb in #7059
• (pydantic-ai) Use FunctionModel for testing output data transformations by @alexander-alderman-webb in #7036
• (serializer) Avoid copying the whole Mapping before trimming to MAX_DATABAG_BREADTH by @Malkiz223 in #6783
• (tests) Replace _transport_works_cases loop with explicit case list by @ericapisani in #6978
• 🤖 Update test matrix with new releases (08/10) by @github-actions in #7142
• Pin litellm<=1.96 for generating test suites by @alexander-alderman-webb in #7147
• Unpin Django version for generating test suites by @alexander-alderman-webb in #7146
• Move Redis breadcrumbs to integration by @sentrivana in #7129
• Move subprocess breadcrumbs to integration by @sentrivana in #7128
• Avoid scanning installed packages when MCP SDK is not installed by @mangodxd in #6865
• Remove test-lambda-locally by @alexander-alderman-webb in #7069
• Add pydantic-ai-slim to uv typing group by @alexander-alderman-webb in #7025
• Add langchain-base to uv typing group by @alexander-alderman-webb in #7016
• 🤖 Update test matrix with new releases (08/03) by @github-actions in #6985
• Mock attributes in redis cluster tests by @sentrivana in #7022
• Deprecate ai_track() by @alexander-alderman-webb in #7005
• Reduce common test suite while not reducing coverage by @ericapisani in #6972
• Use top-level options if available by @sentrivana in #6959
• Restore wrapped functions in functions_to_trace tes...

2.66.1

Bug Fixes 🐛

Tracing

• Handle exceptions raised within traces_sampler and other callbacks by @ericapisani in #6853

Internal Changes 🔧

• Fix flaky decorator test by @sl0thentr0py in #6830

2.66.0

New Features ✨

• (tracing) Promote trace_lifecycle and ignore_spans to top-level options by @ericapisani in #6821

Bug Fixes 🐛

Tracing

• Skip child span creation in streaming path when no current span (HTTP clients) by @sentrivana in #6811
• Skip child span creation in streaming path when no current span (task queues) by @sentrivana in #6814
• Skip child span creation in streaming path when no current span (misc) by @sentrivana in #6815
• Skip child span creation in streaming path when no current span (databases) by @sentrivana in #6808
• Skip child span creation in streaming path when no current span (web frameworks) by @sentrivana in #6810
• Skip child span creation in streaming path when no current span (django) by @sentrivana in #6809

Internal Changes 🔧

• (logging) Fix flaky test_logging_captured_warnings by @ericapisani in #6824
• Remove flaky no cyclic garbage test by @alexander-alderman-webb in #6817
• Remove Claude permission settings by @cleptric in #6806
• Move nullcontext to sentry_sdk.utils by @sentrivana in #6805