The Agent Skills Directory

[SAFE]: The skill utilizes the official ax CLI tool (arize-ax-cli) for interacting with the Arize AI platform, a well-known service in the machine learning observability space.
[SAFE]: Best practices for secret management are enforced. The instructions explicitly forbid reading .env files or searching for credentials on the filesystem, instead directing the agent to use environment variables (ARIZE_API_KEY) and official profile management commands.
[SAFE]: The skill implements an explicit security guardrail regarding untrusted content. It warns that trace data may contain prompt injections and instructs the agent to treat exported data as raw text only, never as instructions to be executed.
[SAFE]: Installation instructions for the ax CLI point to official package registries (PyPI) and use standard, verifiable tools like uv, pipx, and pip.
[SAFE]: Network activity is restricted to legitimate Arize infrastructure (e.g., api.arize.com, flight.arize.com).

arize-trace