The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill interacts with external Google Sheets data, creating a surface for indirect prompt injection.
Ingestion points: Tools such as GOOGLESHEETS_BATCH_GET, GOOGLESHEETS_VALUES_GET, and GOOGLESHEETS_LOOKUP_SPREADSHEET_ROW read content from spreadsheets into the agent's context (SKILL.md).
Boundary markers: The instructions lack specific delimiters or instructions for the agent to ignore potentially malicious embedded content within the retrieved sheet data.
Capability inventory: The skill provides extensive write and management capabilities (e.g., GOOGLESHEETS_BATCH_UPDATE, GOOGLESHEETS_UPSERT_ROWS, GOOGLESHEETS_DELETE_DIMENSION), which could be triggered or abused by malicious data (SKILL.md).
Sanitization: There are no requirements or methods described for sanitizing data fetched from the spreadsheets.
[SAFE]: The skill references tools and documentation from Composio (composio.dev), a well-known platform for AI agent toolkits, and an MCP server endpoint at rube.app. These references are standard for the skill's stated purpose.

googlesheets-automation