Datadog SDK

This release introduces several new features including AI Guard redaction, enhanced LLM Observability, and agentless OpenFeature provider delivery. It also includes numerous bug fixes across various integrations like Botocore, Django, and Tracing.

This release addresses a bug where a large number of view or traced functions could cause exceptions. The fix ensures smoother operation for code origin span handling.

This release addresses a bug where a large number of view or traced functions could cause exceptions. The fix improves stability by ensuring proper handling of code origins for spans.

This release addresses a bug in tracing by preventing a reentrant SpanData borrow panic. The fix ensures more stable operation of tracing functionalities.

This release fixes two bugs: one related to snowflake cursor calls with keyword arguments and another involving a rare crash with gevent during profiling.

This release addresses two specific bugs: one related to snowflake cursor calls with keyword arguments and another concerning a rare crash with gevent and profiling.

This release addresses several critical bug fixes across various integrations and internal components, including memory leak resolutions and stability improvements for Python versions 3.9-3.12.3. It also resolves issues with span export, trace data handling, and LLM observability.

This release addresses several memory growth issues related to profiling and symbol database handling. It also resolves a span generation problem with OpenRouter requests when the OpenAI integration is active.

This release addresses several critical bug fixes related to application crashes in specific Python versions and potential memory growth issues during stack profiling.

This release enhances LLM Observability with features like prompt management, image support, and OpenAI Realtime API instrumentation. It also introduces safer ID generation in Lambda, OpenTelemetry integration, and improved exception profiling.

This release addresses several critical bugs, including issues with the openfeature provider, application crashes on specific Python versions during thread cleanup, and a memory leak related to runtime function wrapping.

This release addresses a critical bug fix related to uwsgi worker crashes during SIGTERM exits. It ensures more stable operation when the application is terminated.

This release includes bug fixes for CI Visibility, addressing an issue with skipped test details in the Datadog UI, and resolves crashes in uwsgi workers during SIGTERM exits.

This release includes a bug fix for the `aws_durable_execution_sdk_python` that corrects the reporting of replayed durable operations that have failed permanently.

This release addresses two bug fixes related to LLM Observability. It resolves a TypeError in streamed tool-call reconstruction and corrects cost reporting for OpenRouter requests with specific integrations.

This release addresses a critical bug fix related to uwsgi worker crashes during SIGTERM exits.

This release includes several bug fixes across CI Visibility, Code Security (IAST), and LLM Observability. It addresses issues with test skipping details, taint tracking, tool-call reconstruction, uwsgi worker crashes, and cost reporting for specific LLM integrations.

This release addresses a potential memory leak in span code origin and fixes an issue with LangChain integration submitting empty chat model stream spans.

This release introduces significant enhancements to LLM Observability, including prompt management, image attachment, and OpenAI Realtime API instrumentation. It also brings improvements to tracing safety, OpenTelemetry integration, AI Guard features, and crash tracking.

This release includes several bug fixes across CI Visibility, Code Security (IAST), MCP, and IAST. Notably, issues related to Intelligent Test Runner correlation IDs, taint mapping in IAST, MCP initialization with LLM Observability, and uvloop compatibility have been addressed.

This release addresses several bugs across CI Visibility, Code Security (IAST), MCP, Code Origin, and LLM Observability. Key fixes include resolving missing test information in Datadog UI, improving taint mapping in IAST, ensuring MCP instrumentation initialization, fixing deepcopying issues, preventing IAST crashes with uvloop, and restoring cost data for OpenAI/LiteLLM integrations with OpenRouter.

This release introduces significant enhancements to AI Guard, LLM Observability, and AppSec, including new features for handling audio, improving evaluation, and detecting LLM usage. It also brings instrumentation support for Ray Serve and Mistral AI, along with profiling improvements and bug fixes.

This release includes bug fixes for PyYAML consumers like Airflow and addresses a false positive vulnerability reporting issue in IAST. Queries in IAST are now scoped to the calling request's slot.