Sentry Instrumentation Guide

You're adding Sentry instrumentation and can't decide if something should be an error, span, log, or metric. This skill is the decision framework: errors need stack traces and get tracked as Issues, traces show where time went in the request waterfall, logs capture what was true at one decision point, and metrics let you chart trends and alert on rates. It walks through the overlaps (span attribute vs metric, log vs span) with a tiebreaker table, then shows a fully instrumented request handler in Python and TypeScript. Decides what to emit, not how to wire it up. Hand off to the language specific SDK skills for actual implementation once you know which signal you need.

skills.sh installs: 1.2K
GitHub stars: 277
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add getsentry/sentry-for-ai --skill sentry-instrumentation-guide --agent claude-code

Installs into .claude/skills of the current project.

Files

All Skills > Feature Setup > Instrumentation Guide

Sentry Instrumentation Guide: When to Reach for What

Errors, traces, logs, and metrics are the four kinds of telemetry most apps run on, and they overlap enough that the choice is rarely obvious. You can stuff context into a span attribute instead of logging it. You can count log lines instead of emitting a metric. You can add a duration to a log and call it a span.

But each signal exists because it answers a different question and feeds a different workflow once it lands. Reaching for the wrong one means the data is technically there but useless for the job you actually have later. This skill is the decision framework: given a value or an event in front of you, which signal should carry it, and why.

It decides what to emit. For how to turn each pillar on for a given stack, hand off to the sentry-*-sdk skills and sentry-setup-ai-monitoring.