The Agent Skills Directory

[COMMAND_EXECUTION]: The skill interacts with the host system by executing resend CLI subcommands (send, list, get, doctor, whoami) to facilitate email operations.
[PROMPT_INJECTION]: The skill processes user-supplied data (email recipients, subjects, and bodies) that is passed to shell commands, creating a potential surface for indirect prompt injection. However, the skill explicitly provides safety guidelines and commands like --text-file - to mitigate shell injection and quoting issues. * Ingestion points: User-provided values for email to, subject, and body fields in SKILL.md. * Boundary markers: The skill does not use specific delimiters for user input but instructs the agent to use standard input for body content to avoid shell issues. * Capability inventory: Shell execution of the resend binary for email delivery and management. * Sanitization: The instructions mandate manual confirmation of all email fields and dry-run testing for complex content.

resend-email-cli