Arize Prompt Optimization

This pulls prompts from your production traces, correlates them with eval scores and human annotations, and runs an optimization loop using Arize's ax CLI. You point it at LLM spans in your project, it extracts structured chat messages or prompt templates, then pairs them with feedback signals like correctness labels or LLM-as-judge explanations. The workflow is thorough: export spans, reconstruct the prompt as a messages array, merge performance data from datasets and experiments, then iterate. Most useful when you're already logging OpenInference traces to Arize and have eval or annotation data sitting there. It won't guess at credentials or search your filesystem, it uses ax profiles and ai-integrations commands, which keeps things clean if you're working in a locked-down environment.

skills.sh installs: 1.1K
GitHub stars: 41K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add github/awesome-copilot --skill arize-prompt-optimization --agent claude-code

Installs into .claude/skills of the current project.

Files

Arize Prompt Optimization Skill

SPACE — All --space flags and the ARIZE_SPACE env var accept a space name (e.g., my-workspace) or a base64 space ID (e.g., U3BhY2U6...). Find yours with ax spaces list.

Concepts

Where Prompts Live in Trace Data

LLM applications emit spans following OpenInference semantic conventions. Prompts are stored in different span attributes depending on the span kind and instrumentation:

SKILL.md on GitHub

Sponsors

Promote your product hereReach 125k+ visitors monthly Promote your product hereReach 125k+ visitors monthly