The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process email threads from external sources, which creates a potential surface for indirect prompt injection attacks. Any email content could contain malicious instructions designed to trick the agent into performing unintended actions.
Ingestion points: The workflow in SKILL.md instructs the agent to read existing email threads before composing replies or triaging messages.
Boundary markers: There are no explicit instructions or delimiters defined to help the agent distinguish between user instructions and potentially malicious content embedded within the email threads.
Capability inventory: The skill has the capability to draft, reply to, and send emails through the mail surface.
Sanitization: No specific sanitization or filtering of email content is mentioned to prevent the agent from following instructions found within the body of an email.

email-ops