The Agent Skills Directory

[SAFE]: The skill demonstrates proactive security measures by explicitly instructing the agent to never request authentication tokens directly in the chat interface and never to echo them in logs or responses.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute dino CLI commands for checking authentication status and logging out. These operations are limited to the specific functionality of the Dinox tool.
[DATA_EXFILTRATION]: No patterns of unauthorized data transmission were found. The skill includes specific safeguards to prevent the accidental exposure of the DINOX_TOKEN environment variable or saved credentials.
[CREDENTIALS_UNSAFE]: No hardcoded credentials or secrets were detected. The skill correctly identifies that tokens should be managed by the user in their local environment rather than handled by the AI agent.

dino-auth