Arize-aiArize-aiOfficialยท5 Agent Skills Included

openinference

Trace, debug, and review AI applications across Python, JS, and Java

Adds OpenTelemetry-based tracing to AI apps built with OpenAI, LangChain, LlamaIndex, Anthropic, and dozens of other frameworks. Captures LLM calls, token usage, tool invocations, and agent sessions automatically without manual logging code. Includes agent skills for debugging traces with the Phoenix CLI, reviewing instrumentation code, and fixing CI failures.
npx skills add Arize-ai/openinference --all -g -y
Available:

Tells the AI agent how the monorepo is organized, which build and test commands to run per language, and the rules every new instrumentation package must follow.

All Skills in This Repository (5)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install OpenInference?โ–ผ

Run `npx skills add Arize-ai/openinference --all -g -y` in your terminal to install all skills in this suite globally.

What is OpenInference used for?โ–ผ

It adds automatic tracing to AI applications so you can see every LLM call, token count, and tool invocation. It works with any OpenTelemetry-compatible backend like Arize Phoenix.

Which AI frameworks does OpenInference support?โ–ผ

It ships ready-made instrumentors for OpenAI, Anthropic, LangChain, LlamaIndex, DSPy, AWS Bedrock, CrewAI, and many more across Python, JavaScript, Java, and Go.

How do I debug failing LLM traces with AI?โ–ผ

The included phoenix-cli skill lets your AI agent fetch traces, annotate errors, and group failures into categories using simple terminal commands.

Does OpenInference work with Claude Code and other AI agents?โ–ผ

Yes. The bundled skills follow the standard SKILL.md format and run in Claude Code, Cursor, Codex, and other compatible agents.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’