sf-ai-agentforce-grid

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly reads org-hosted, user-authored Grid content (e.g., worksheet cells, prompt templates, agent outputs) via the Grid REST endpoints and helper scripts (see SKILL.md's guidance to read /worksheets/{id}/data and the scripts worksheet_to_rows.mjs, grid_api_request.mjs, and grid_smoke_test.mjs), so untrusted third-party Salesforce org content is ingested and can influence subsequent actions.