The Agent Skills Directory

[SAFE]: The skill follows security best practices for credential management. It explicitly instructs the agent to avoid reading sensitive files like .env and instead utilizes environment variables and official CLI profile management (ax profiles).- [SAFE]: The skill references the arize-ax-cli package, which is an official tool provided by the vendor (Arize-ai) for interacting with their platform. Installation instructions use standard package managers like pip, pipx, and uv.- [SAFE]: Python code snippets included in the skill are simple one-liners used for formatting or inspecting local JSON data exported by the CLI. These scripts do not perform network operations or access sensitive system resources.- [SAFE]: Instructions for handling user API keys are secure, emphasizing that keys should never be pasted into the chat or logged, and providing links to official administration pages for key retrieval.

arize-evaluator