google-workspace

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [NO_CODE]: The provided skill package consists exclusively of documentation and metadata files. No source code or executable scripts (such as Python or JavaScript files) are present to implement the described functionality.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it processes untrusted content from external sources.
  • Ingestion points: Content is retrieved from external sources via read_sheets_data (references/sheets.md), get_docs_content (references/docs.md), and list_calendar_events (references/calendar.md).
  • Boundary markers: There are no defined boundary markers or instructions to the agent to ignore potentially malicious instructions embedded within the retrieved data.
  • Capability inventory: The skill possesses significant capabilities including writing data (write_sheets_data), modifying documents (insert_docs_content), sharing files (share_drive_file), and deleting resources (delete_drive_file).
  • Sanitization: No evidence of sanitization or validation of the retrieved content is present in the documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 02:43 PM
Security Audit — agent-trust-hub — google-workspace