The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Gmail threads and searches which may contain malicious instructions.
Ingestion points: gmcli search and gmcli thread commands in SKILL.md ingest external email content into the agent's context.
Boundary markers: None present to delimit untrusted email content from agent instructions.
Capability inventory: The skill possesses the capability to send emails and list drafts via gmcli send and gmcli drafts list commands.
Sanitization: No specific sanitization or filtering of email body content is mentioned.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of a global NPM package from the author's registry.
Evidence: npm install -g @mariozechner/gmcli in SKILL.md.

gmcli