
Arize Annotation Manager
Set up label schemas and human review queues on Arize and bulk-apply annotations to spans.
- Repository
- github/awesome-copilot
- Stars
- 40,703
- Licence
- MIT
- Files
- 3 files
What it does
This skill teaches an AI agent how to manage human feedback on Arize, a platform for monitoring AI apps. The agent can create annotation configs (the label definitions, such as correct/incorrect, a 0-10 score, or free-text notes), set up annotation queues that route records to human reviewers, and apply annotations to project spans through the Python SDK. It relies on the ax command-line tool and a configured Arize profile to act on your account.
Great for
- ML or LLM engineers who run evaluation and labeling workflows on Arize
- Data teams organising human review of AI outputs
- Teams that need to script bulk annotation of traced spans
Skip it if
- Users without an Arize account or the ax CLI, since nothing works without them
- Anyone wanting to label data outside Arize, because the skill only targets Arize spaces, configs and queues
- Users who might paste API keys into chat: the skill tells the agent to use ax profiles instead, but credentials still pass through the agent's session
Try asking
- “Create a categorical annotation config called Correctness with correct and incorrect values”
- “Set up an annotation queue for reviewers to label a support-bot project, assigned to two emails”
- “Bulk-apply human annotations to spans in my project using update_annotations”
- “List the records waiting in my Correctness Review queue”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 3 original files, the MIT licence and an attribution note.
arize-annotation.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip arize-annotation.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip arize-annotation.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Create the annotation config in the space before expecting labels to persist, since queues and labels depend on it
- Remember that list flags such as --annotation-config-id fully replace existing values on update, so pass every desired value
- Deleting a config is irreversible, so confirm before using --force
What's inside
- references/ax-profiles.md
- references/ax-setup.md
- SKILL.md
Redistributed unmodified from github/awesome-copilot under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.