The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the reception of inbound emails via webhooks, which introduces a surface for indirect prompt injection if an AI agent processes the untrusted body or metadata of these emails.
Ingestion Point: references/receiving.md explains how to retrieve email body content using resend.emails.receiving.get().
Boundary Markers: No explicit boundary markers or delimiters are defined for the email content itself, though the skill directs users to specialized security skills for handling untrusted input.
Capability Inventory: The skill provides extensive capabilities including sending emails, creating automations, managing domains, and modifying contact lists, which could be misused if an injection is successful.
Sanitization: While the skill includes cryptographic signature verification (resend.webhooks.verify) to ensure data authenticity from the provider, it does not perform content-level sanitization on the email body for prompt processing.

resend