google-calendar-automation

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to configure an external MCP server endpoint at https://rube.app/mcp for tool execution.
  • [DATA_EXFILTRATION]: Sensitive calendar information, including event details and attendee email addresses, is processed and transmitted through the external rube.app infrastructure.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection via external data sources.
  • Ingestion points: Malicious instructions can enter the agent context through event summaries, descriptions, and location fields retrieved via GOOGLECALENDAR_FIND_EVENT and GOOGLECALENDAR_EVENTS_LIST (SKILL.md).
  • Boundary markers: The instructions lack boundary markers or explicit warnings for the agent to ignore instructions embedded within retrieved calendar data.
  • Capability inventory: The skill allows for significant state changes, including GOOGLECALENDAR_CREATE_EVENT, GOOGLECALENDAR_PATCH_EVENT, and GOOGLECALENDAR_DELETE_EVENT (SKILL.md).
  • Sanitization: No sanitization or validation routines are described for data fetched from the Google Calendar API before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 09:12 AM
Security Audit — agent-trust-hub — google-calendar-automation