The Agent Skills Directory

[SAFE]: The skill serves as an interface for the ShellGames.ai platform. All documented behaviors, including registration, authentication via JWT, and game state management, are consistent with the skill's stated purpose as a gaming integration.
[DATA_EXPOSURE]: The skill documentation includes examples for handling sensitive data such as user passwords and authentication tokens. These are appropriately presented as placeholders (e.g., your-secure-password, jwt...) and follow industry-standard practices for secure API interactions.
[COMMAND_EXECUTION]: The instructions suggest using well-known networking tools like cloudflared or ngrok for exposing local webhooks to receive 'wake' notifications. This is a common and legitimate practice for developers integrating with external services that require callback URLs.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates interaction with untrusted data through in-game chat and direct messaging systems. While this represents a theoretical attack surface common to all messaging integrations, the skill does not grant elevated privileges or perform unsafe operations based on the content of these messages.

shellgames