google-sheets

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly reads and writes arbitrary Google Sheets specified by --sheet-id / SHEET_ID (see SKILL.md and scripts/cli.py open_sheet / cmd_read / cmd_append), meaning the agent will ingest user-generated/untrusted spreadsheet content from third-party Google Sheets which could influence its subsequent actions.