The Agent Skills Directory

[SAFE]: The skill implements robust credential safety guidelines. It explicitly forbids the agent from reading .env files or searching the filesystem for secrets, instead directing the use of official CLI tools (ax profiles) and environment variables.
[SAFE]: Instructions for handling API keys follow industry best practices. The skill documentation emphasizes never passing raw secrets as command-line arguments and never asking users to paste keys directly into the chat interface.
[SAFE]: All identified software dependencies, including the arize Python SDK, arize-ax-cli, and pandas, are standard, reputable libraries provided by or associated with the vendor and required for the stated functionality.
[SAFE]: No evidence of prompt injection, code obfuscation, unauthorized data exfiltration, or persistence mechanisms was found in the instructions or reference materials.

arize-annotation