The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a system binary named gws to perform Gmail API operations. It provides a structured interface for the agent to execute commands like gws gmail users.messages.get or gws gmail users.threads.list to manage mailbox data.
[INDIRECT_PROMPT_INJECTION]: As a mail management tool, the skill is designed to ingest and process untrusted data from external Gmail messages, which represents a potential attack surface for indirect prompt injection.
Ingestion points: Functions such as +read, +triage, and users.messages.get are used to retrieve the body and headers of incoming emails (SKILL.md).
Boundary markers: The current instructions do not specify the use of delimiters or 'ignore' instructions to isolate processed email content from the agent's primary directives.
Capability inventory: The skill grants the agent the ability to send, forward, and reply to emails, which could be leveraged if the agent inadvertently follows instructions embedded in a message.
Sanitization: There are no explicit instructions for the agent to sanitize or validate the content of the emails before processing or responding to them.

gws-gmail