Arize Annotation

If you're running human review workflows on Arize, you need annotation configs and queues set up before anyone can label anything. This skill handles both: creating label schemas (categorical, continuous, or freeform), managing review queues, and bulk-updating spans with human feedback via the Python SDK. The CLI covers all the CRUD operations, and the docs are explicit about which surfaces labels apply to (spans, dataset examples, experiment outputs, queue items). One thing to note: list flags like annotation-config-id fully replace existing values on update, so you need to pass everything back, not just what changed. Useful when you're setting up labeling pipelines or need programmatic access to review workflows.

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-annotation --agent claude-code

Installs into .claude/skills of the current project.

Files

Arize Annotation 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.

This skill covers annotation configs (the label schema) and annotation queues (human review workflows), as well as programmatically annotating project spans via the Python SDK.

Direction: Human labeling in Arize attaches values defined by configs to spans, dataset examples, experiment-related records, and queue items in the product UI. This skill covers: ax annotation-configs, ax annotation-queues, and bulk span updates with ArizeClient.spans.update_annotations.