ClawHub
Back to skill

Security audit

Email Triage

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent Gmail triage skill that reads recent email only to summarize actionable items, with no executable code, persistence, or hidden data handling found.

Install only if you are comfortable letting Claude read recent Gmail messages for the requested window. Use an explicit invocation like /email-triage and review any drafted replies before taking further action.

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 (3)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The recommended mobile trigger "Check my emails" is a common natural-language phrase that can easily be spoken or typed in ordinary conversation, increasing the chance of unintended skill activation. Because this skill accesses Gmail and performs inbox triage, accidental invocation could expose sensitive message metadata and summaries when the user did not explicitly intend to run the skill.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The example trigger "/email-triage" is safe, but the finding at this line corresponds to a broad everyday phrase in the examples section that overlaps with normal user speech. In a skill that reads Gmail content, broad trigger wording raises the risk of accidental activation and unintended access to or display of sensitive inbox information.

Vague Triggers

Medium
Confidence
96% confidence
Finding
Several example triggers are phrased as ordinary requests like "What emails need my attention?" and "Any urgent emails I need to deal with?" without requiring explicit invocation. This makes accidental or implicit activation more likely, and in the context of a Gmail-connected skill, that can lead to unnecessary retrieval, classification, and summarization of potentially sensitive emails.

Static analysis

No suspicious patterns detected.