arize-instrumentation

An Arize skill that optimizes, improves, and debugs LLM prompts using production trace data, evaluations, and annotations. Runs data-driven optimization loops via the ax CLI.

An observability skill from Arize's 11-skill collection for LLM tracing, instrumentation, datasets, experiments, and evaluators. Requires the ax CLI for authentication and integrates with 40+ AI coding agents including Cursor, Claude Code, and Codex.

Skills for AI coding agents to manage datasets, add observability, and run experiments on LLM applications using the Arize platform, with workflows for tracing, debugging, and prompt optimization in production.

Skills that guide AI coding agents to add observability, run experiments, and optimize prompts for LLM applications using the Arize platform and the ax CLI, with support for tracing, debugging, and production monitoring workflows.

An observability skill from Arize's 11-skill collection for AI provider integration and LLM optimization. Requires the ax CLI for authentication and integrates with 40+ AI coding agents including Cursor, Claude Code, and Codex.

Arize skill for creating **LLM-as-judge evaluators**, running evaluation tasks, and setting up continuous monitoring — part of the Arize platform's skills that guide AI coding agents to add observability, run experiments, and optimize prompts for LLM applications using the `ax` CLI.

Create and manage LLM provider credentials (OpenAI, Anthropic, Azure, Bedrock, Vertex, and more).

Create and manage annotation configs (categorical, continuous, freeform); bulk-annotate project spans via the Python SDK.

Audit an AI agent for regulatory compliance against frameworks including the EU AI Act, NIST AI RMF, GDPR, and HIPAA, producing a tailored remediation checklist. Part of the Arize observability skills suite.

Agent skills that guide AI coding agents to add LLM observability, run experiments, and optimize prompts for LLM applications using the Arize platform. The skills encode the Arize team's refined workflows for debugging LLM apps in production, handling CLI flags and multi-step recipes automatically.