memory-email-ingest

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.95). The skill reads full email bodies via the Gmail MCP tools (gmail_read_message for each messageId returned by gmail_search_messages), and those email contents are authored by other people (calendar contacts), so outsider free text is ingested into the agent’s LLM context.