skills.rest
handsomelong922
handsomelong922COMMUNITY
Explore all skills by this author →
GitHub

phoenix-observability

Collect traces and evaluate LLM outputs with OpenTelemetry instrumentation.

Updated Apr 3, 2026
Search Tags:#monitoring#llm#evaluation#observability#tracing#opentelemetry#self-hosted
One-click install
npx skills add https://github.com/handsomelong922/my-codex-skills --skill phoenix-observability-handsomelong922
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: phoenix-observability
Source: https://github.com/handsomelong922/my-codex-skills/tree/main/skills/phoenix
Command: npx skills add https://github.com/handsomelong922/my-codex-skills --skill phoenix-observability-handsomelong922

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Phoenix Observability provides a self-hosted platform to trace, evaluate, and monitor LLM pipelines, enabling you to diagnose issues and measure performance without vendor lock-in.

Core Features & Use Cases

  • Tracing and instrumentation with OpenTelemetry across LLM workflows for end-to-end visibility.
  • Built-in evaluation framework with datasets and experiments to assess model quality in production.
  • Real-time monitoring and dashboards for production AI systems with privacy and control.

Quick Start

Install arize-phoenix and start a local server to begin collecting traces and evaluating models.

Frequently Asked Questions about phoenix-observability

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I trace LLM applications using OpenTelemetry for observability?▼

You can trace LLM applications by applying OpenTelemetry instrumentation across your workflows to gain end-to-end visibility into model pipelines and diagnose issues without vendor lock-in.

What is the best way to evaluate LLM model quality in production?▼

Evaluating LLM model quality in production requires an integrated evaluation framework that leverages datasets and experiments to assess outputs and measure performance continuously.

Can I monitor production AI systems with a self-hosted observability platform?▼

Yes, you can deploy a self-hosted observability platform to monitor production AI systems in real-time using dashboards, ensuring data privacy and control without relying on external vendors.

Do I need PostgreSQL or SQLite to run a self-hosted LLM tracing server?▼

You need either PostgreSQL or SQLite as optional backends to support a self-hosted deployment for collecting traces and evaluating model outputs within your local environment.

How do I debug LLM workflows when model outputs are unexpected?▼

Debugging LLM workflows involves collecting end-to-end traces and running evaluations against datasets to identify performance bottlenecks and diagnose issues within your application pipelines.

Related Skills

Explore More Software Engineering Skills→
Clay-HHKClay-HHK
2

phoenix-observability

Collect and analyze LLM execution traces and evaluation results via OpenTelemetry.

Community
Advanced
KappTech88KappTech88

langsmith-observability

Diagnose LLM behavior across tracing, evaluation, and monitoring workflows in LangSmith.

Community
Advanced
petrkindlmannpetrkindlmann
64

observability-driven-testing

Convert production traces, logs, and telemetry into actionable tests.

Community
Advanced
gulati8gulati8
3

infrastructure-patterns

Evaluate infrastructure patterns using cost, observability, security, and deployment topology frameworks.

Community
Advanced