Back to skill

Security audit

Google calender

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward Google Calendar gateway guide, but it enables reading, changing, deleting, and notifying people from the user's primary calendar without clear confirmation guidance.

Review this skill before installing if you use Google Calendar for sensitive personal or business scheduling. Only use it with a Maton API key you trust, confirm event IDs before updates or deletion, and explicitly approve attendee email addresses and whether notifications should be sent.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (5)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The document emphasizes Australia/Melbourne and uses that time zone throughout the examples, which can function as a locale constraint. Because no opt-in or justification is provided, this appears to force a specific locale contrary to the natural-language policy requirement.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 45)May include surrounding context.

📅 1. List Events

bash
curl "https://gateway.maton.ai/google-calendar/calendar/v3/calendars/primary/events?maxResults=20&singleEvents=true&orderBy=startTime" \
-H "Authorization: Bearer $MATON_API_KEY"

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The example sends invitations/updates to attendees using sendUpdates="all" without warning that this triggers outbound communications to external recipients. An agent following this skill could contact unintended parties, leak meeting metadata, or create unauthorized invitations without clear user consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

Updating attendees with sendUpdates="all" can immediately notify recipients, but the skill provides no caution about external communications or consent requirements. This can cause privacy issues, unintended outreach, and unauthorized calendar changes visible to third parties.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill documents a destructive delete operation with no caution, confirmation requirement, or note that deletion can permanently remove calendar events. In an agent context, this increases the risk of accidental or unauthorized destructive actions being performed on a user's primary calendar.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.