The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external email senders that can be used to influence agent behavior.
Ingestion points: The listEmails, readEmail, and searchEmails methods in handler.ts ingest email subjects, snippets, and full message bodies into the agent's context.
Boundary markers: The skill does not use delimiters or provide instructions for the agent to ignore commands found within email bodies.
Capability inventory: The skill possesses capabilities like send_email, reply_to_email, and archive_email. Instructions also mention workspace_write_file and send_notification.
Sanitization: No escaping or filtering is performed on the ingested email content.

google-gmail