The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the installation of the agentmail Python package from a public registry to provide its core functionality.
[COMMAND_EXECUTION]: The skill utilizes shell commands for setup (mkdir, chmod, cat) and provides Python scripts (check_inbox.py, send_email.py) to interact with the email service. These are used for legitimate configuration and operation of the email client.
[DATA_EXPOSURE]: The skill manages an API key stored in ~/.agentmail/config.json. It follows security best practices by recommending chmod 600 permissions to restrict access to the current user.
[INDIRECT_PROMPT_INJECTION]: As the skill is designed to read incoming emails from external senders, it introduces a surface for indirect prompt injection. The documentation explicitly acknowledges this risk, advising users to treat incoming content as untrusted and to avoid automated actions on sensitive data without human approval.

agentmail