The Agent Skills Directory

[PROMPT_INJECTION]: The skill provides workflows for reading and searching content from external Google Sheets using tools like GOOGLESHEETS_BATCH_GET and GOOGLESHEETS_LOOKUP_SPREADSHEET_ROW. This data constitutes an untrusted ingestion point where malicious instructions could be embedded in spreadsheet cells to influence agent behavior. The instructions do not specify boundary markers or data sanitization protocols for the ingested content.
[SAFE]: The skill references the Rube MCP endpoint (rube.app) and Composio documentation (composio.dev) for legitimate configuration purposes. These references are directed at well-known services associated with the skill's primary functionality.

googlesheets-automation