ClawHub
Back to skill

Security audit

Email Triage

Security checks for vulnerabilities and agentic risk

Overview

This Gmail triage skill reads recent inbox content to summarize actionable emails, which fits its stated purpose and has no hidden execution, persistence, or exfiltration behavior.

Install only if you are comfortable with Claude using the Gmail connector to read recent email metadata and message body excerpts for triage. For tighter control, invoke it with an explicit time window and ignore/include rules, such as only the last 8 hours or only selected senders.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill instructs Claude to fetch sender details, subjects, timestamps, and email bodies, but the skill description and user-facing flow do not prominently warn that private email content will be read. This creates a consent and privacy risk: users may trigger the skill expecting a lightweight summary without realizing the agent will access message contents, including potentially sensitive financial or personal communications.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrases include very broad everyday language such as "Check my emails" and "What emails need my attention?", which can cause the skill to activate in situations where the user did not intend a full inbox triage. Because this skill reads email metadata and message bodies, accidental invocation can expose sensitive personal or business content beyond the user’s immediate intent.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.