The Agent Skills Directory

[SAFE] (SAFE): The skill is purely instructional and defines best practices for real-time systems architecture. It does not contain executable scripts or code that interacts with the filesystem or network.- [CREDENTIALS_UNSAFE] (SAFE): The validation rules in references/validations.md actively discourage unsafe credential handling by flagging hardcoded URLs and authentication tokens sent within message payloads.- [DATA_EXFILTRATION] (SAFE): No sensitive file paths or exfiltration vectors were detected. The skill focuses on standard communication protocols.- [PROMPT_INJECTION] (SAFE): The skill uses standard identity and role-play definitions without attempting to bypass safety guardrails or override system instructions.- [COMMAND_EXECUTION] (SAFE): No subprocess calls or shell commands are present in the provided files.

websockets-realtime