Packages
Captures and visualizes the internal execution of LLM application flows (DAG and Flex) with OpenTelemetry-compatible tracing for observability.
Auto-instruments DSPy applications to generate OpenTelemetry-compatible traces for observability and debugging, sending them to collectors like Arize Phoenix or Arize AX.
Routes Python logging records to AWS S3 and Kinesis using multithreaded handlers with optional gzip compression for S3.
Measures and tracks carbon emissions from local computing hardware (CPU, GPU, RAM) and provides estimates of their environmental impact based on regional electricity carbon intensity.
Install it if environmental impact of your code matters to your workflow or reporting requirements.
SNMP Simulator creates a pure-Python SNMP agent simulator that responds to SNMP queries (v1/v2c/v3) as if it were multiple real network devices, using configurable data files or plugins.
Install it if you need to test against SNMP agents without physical devices or if you're developing SNMP client code.
Integrates Cohere's language model APIs into Apache Airflow workflows as operators and hooks for task orchestration.
Integrates pgvector operations into Apache Airflow workflows, providing operators and hooks to manage vector embeddings in PostgreSQL.
django-watchman exposes a status endpoint that reports the health of your application's backing services—databases, caches, storage—as JSON, enabling external monitoring and introspection.
Install it if you need external monitoring of your application's backing services.
Integrates Qdrant vector database with Apache Airflow, enabling vector search and similarity operations within Airflow workflows and DAGs.
Install only if you have Airflow >=2.11.0 and a use case that requires vector search within your DAGs.
Integrates Apache Airflow with DingTalk, enabling workflow notifications and messaging to DingTalk channels from Airflow DAGs.
PyMonCtl provides cross-platform Python access to monitor information and control—querying resolution, position, brightness, and orientation, and setting display modes and geometry across Windows, Linux, and macOS.
Integrates Apprise notification delivery into Apache Airflow, enabling tasks to send alerts and messages through multiple notification channels via a unified provider interface.
Install it if you run Airflow and need to send notifications through Apprise-supported channels.
Safe Init wraps AWS Lambda handlers to provide automatic error handling, execution tracing, and integration with monitoring services like Sentry, Datadog, and Slack.
Integrates Apache Airflow with PowerShell Remoting Protocol (PSRP) to enable remote command execution on Windows systems as part of Airflow workflows.
Integrates Apache Pinot with Apache Airflow, providing operators and hooks to query and manage Pinot clusters as part of Airflow data pipelines.
Install it if you run Airflow and need to query or manage Apache Pinot clusters.
Exposes CloudWatch metrics, logs, alarms, and PromQL queries through a Model Context Protocol (MCP) server that AI agents can use for root cause analysis and troubleshooting recommendations.
Integrates Apache Pig with Apache Airflow, providing operators and hooks to orchestrate Pig scripts as Airflow tasks within data pipelines.
Provides a pure-Python interface to Linux's prctl() system call and Linux capabilities using ctypes, allowing direct manipulation of process attributes from Python code.
Provides optional C/Rust-compiled speedups for Prometheus client_python's multiprocess mode, designed to accelerate aggregation of metrics from many .db files.
Integrates ArangoDB with Apache Airflow, providing operators and hooks to query and manage ArangoDB databases within Airflow DAGs.
Auto-instruments MCP tool calls to propagate OpenTelemetry context between the caller and the tool execution, enabling distributed tracing across MCP interactions.
Manages process ID files with stale detection, file locking, and permission control, available as a context manager or decorator for daemon and single-instance enforcement.
Collects stack traces and diagnostic data from jobs running on Cloud TPU, capturing faults and periodic snapshots to help debug hung or failing workloads.
However, verify the license terms before production use, and confirm it meets your TPU environment's requirements.
Honeybadger captures uncaught exceptions in Python applications and sends them to the Honeybadger error-monitoring service, with built-in support for Django, Flask, AWS Lambda, ASGI, and FastAPI.
Install only if you have a Honeybadger account and API key.
Defines CloudWatch metrics, alarms, and dashboards as code using AWS CDK v1 constructs, enabling programmatic monitoring infrastructure for AWS resources.
Install only if you are maintaining a legacy CDK v1 codebase with no near-term migration path.
Provides Apache Airflow integration for Apache Drill, enabling Airflow workflows to connect to and query Drill clusters via SQL operators and hooks.
Install it if you run Airflow and need to query Apache Drill clusters from your DAGs.
Collects and preserves SLURM job efficiency metrics in Parquet format, analyzing CPU and memory usage patterns before SLURM's accounting database purges them after 30 days.
Adds OpenTelemetry tracing to the Valkey Python client, capturing operations and performance metrics for observability.
However, note that it is a community-maintained external package pending official Valkey support in opentelemetry-python-contrib; verify compatibility with your…
Integrates OpenFaaS serverless functions into Apache Airflow workflows, allowing you to invoke OpenFaaS functions as Airflow tasks.
Integrates Apache Kylin OLAP analytics with Apache Airflow workflows, enabling scheduled queries and data cube management through Airflow DAGs.
Integrates Singularity container runtime with Apache Airflow, enabling Airflow workflows to execute tasks inside Singularity containers via the spython library.
Install only if Singularity (or Apptainer) is available on your worker nodes; otherwise it will not function.
Sends time-series metrics to Prometheus-compatible remote storage backends using the Prometheus Remote Write protocol, with support for authentication, batching, and custom headers.
Provides operators and hooks to integrate Apache Airflow with Zendesk, enabling orchestration of Zendesk API interactions within Airflow DAGs.
Detects and registers container runtime metadata (container ID, image name, pod details) as OpenTelemetry resource attributes for traces and metrics.
nassl is an OpenSSL wrapper for Python that exposes low-level SSL/TLS functions needed for SSL server scanning and testing, with a C extension layer that closely mirrors OpenSSL's API.
Queries Minecraft servers (Java, Legacy, and Bedrock editions) to retrieve status, player counts, latency, and server properties via the Server List Ping and Query protocols.
Install it if you need to check Minecraft server status or player information in a Python application.
Adds distributed tracing instrumentation to scikit-learn operations, capturing them as spans in OpenTelemetry.
Install only if you have an OpenTelemetry tracer/exporter already configured to receive spans.
Interrupts long-running Python code by raising an exception after a specified timeout, using signal-based interruption on the main thread.
Adds distributed tracing to FastAPI, gRPC, and NATS services, propagating trace IDs across async boundaries and message brokers.
oslo.reports generates structured debug reports from running OpenStack services, capturing thread state, configuration, stack traces, and package information in JSON, XML, or templated text format.
Not recommended for non-OpenStack projects unless you need a general-purpose runtime state capture framework and are willing to work with its OpenStack-centric design.