What are the 'Four Golden Signals' mentioned in the skill?

The Four Golden Signals are Latency (time to service a request), Traffic (demand placed on the system), Errors (rate of failed requests), and Saturation (how 'full' the service is).

What Python libraries does this skill support?

This skill focuses on industry-standard libraries including structlog for structured logging, prometheus_client for metrics, and OpenTelemetry for distributed tracing.

Why should I use structured logging instead of the standard Python logging module?

Structured logging outputs data as JSON, which is machine-readable. This allows logging platforms like ELK, Datadog, or New Relic to index fields, enabling complex queries and faster troubleshooting.

How does this skill handle request tracing?

It implements correlation ID propagation, where a unique ID is generated at the ingress point and passed through all internal operations and downstream HTTP calls to link logs and spans together.

Is this skill compatible with FastAPI and Flask?

Yes, the provided patterns include middleware and decorator examples specifically designed to work with modern Python web frameworks and asynchronous clients like httpx.

Python Observability Patterns

Name: Python Observability Patterns
Author: wshobson

bywshobson

•

27,605

•

分析と監視

Implements production-grade observability in Python applications through structured logging, Prometheus metrics, and distributed tracing.

The python-observability skill equips Claude with the expertise to instrument Python applications for high-visibility production monitoring. It provides standardized patterns for implementing JSON-based structured logging using structlog, tracking the 'Four Golden Signals' via Prometheus, and managing distributed tracing with OpenTelemetry. By integrating these practices, developers can ensure their systems are debuggable without code changes, enabling faster incident response and deeper performance analysis through correlation IDs and bounded metric cardinality.

主な機能

01Instrumenting the Four Golden Signals: Latency, Traffic, Errors, and Saturation

02OpenTelemetry integration for modern distributed tracing architectures

0327,605 GitHub stars

04Structured JSON logging configuration with structlog for machine-readable logs

05Prometheus metric collection with optimized label cardinality

06Correlation ID propagation patterns for end-to-end request tracing

ユースケース

01Implementing end-to-end request tracing across a complex chain of Python services

02Setting up centralized logging for microservices to enable faster production debugging

03Creating Prometheus dashboards to monitor service latency and error rates in real-time

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add wshobson/agents python-observability

For use in Claude.ai and ChatGPT

Download Skill