gmail-opportunity-progress

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

• Third-party content exposure detected (high risk: 0.95). The skill’s runtime workflow uses Gmail MCP/API to “read … body” of candidate emails and then classifies them, so any email authored by someone other than the operating user (e.g., recruiters/companies) becomes free-form text ingested into the agent’s LLM context via the Gmail message body/snippet.