Arize Instrumentation

This walks you through adding Arize AX tracing to your LLM app in two phases: first it analyzes your codebase to figure out what you're using (LangChain, OpenAI, whatever), then it implements the instrumentation after you confirm the plan. The approach is solid because it doesn't guess or spray changes everywhere. It prioritizes auto-instrumentation over manual spans, won't touch your business logic, and has explicit rules about never hardcoding credentials. Supports Python, TypeScript, Java, and Go with framework-specific routing for everything from LangGraph to Bedrock. If you need observability on an existing app without rewriting half of it, this is the structured way to do it.

skills.sh installs: 3.6K
GitHub stars: 72
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add arize-ai/arize-skills --skill arize-instrumentation --agent claude-code

Installs into .claude/skills of the current project.

Files

Arize Instrumentation Skill

Add Arize AX tracing to an app for the first time: detect the stack → fetch the one matching integration doc → wire auto-instrumentation → verify a trace lands.

Route locally. Map the detected stack to a single doc page via references/integration-routing.md and fetch only that page. When the app uses an agent framework, route on the framework, not the provider SDK it wraps — a bare openai/anthropic import inside a framework app is not the integration target; route to a provider page only when the app calls the provider SDK directly with no framework in play. When the stack isn't listed there, search llms.txt for its integration page per the router; with no match, use manual instrumentation. Never bulk-read the PROMPT.md aggregate.