The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses curl to make HTTP requests to a Rocket.Chat instance. This is a standard and expected method for interacting with web APIs in this context.
[CREDENTIALS_SAFE]: The instructions correctly leverage environment variables ($ROCKETCHAT_AUTH_TOKEN and $ROCKETCHAT_USER_ID) for authentication, which prevents the leakage of sensitive credentials within the skill itself.
[DATA_EXPOSURE_AND_EXFILTRATION]: Communication is directed to a host defined by placeholders ({{ROCKETCHAT_HOST}}), which typically resolves to an internal service or a user-defined instance, posing no inherent risk of unauthorized data exfiltration.

rocketchat-send