The Agent Skills Directory

[COMMAND_EXECUTION]: The skill functions by instructing the agent to execute shell commands using the resend CLI utility. It assumes the tool is present in the environment and configured with the RESEND_API_KEY.- [DATA_EXFILTRATION]: The skill provides a high-capability channel for transmitting local data to external endpoints. The --attachment flag allows the agent to access and upload any local file the agent has permissions to read, while the --to flag enables sending this data to arbitrary email addresses. This creates a risk of sensitive file exfiltration (e.g., .env, SSH keys, or cloud credentials) if the agent is improperly prompted.- [INDIRECT_PROMPT_INJECTION]:
Ingestion points: The skill processes external data by reading file content via the --html-file flag and binary data via the --attachment flag.
Boundary markers: Absent. There are no instructions provided to the agent to distinguish between user-intended content and potentially malicious instructions embedded in files it may be asked to summarize and email.
Capability inventory: The skill combines file system read capabilities with network transmission (email delivery) across all documented scripts.
Sanitization: None. The skill lacks input validation or sanitization mechanisms for the content being sent.

resend-email