ClawHub
Back to skill

Security audit

Gws Gmail Watch

Security checks for vulnerabilities and agentic risk

Overview

This skill documents a Gmail watch command with sensitive but purpose-aligned Gmail and Pub/Sub access, and no hidden or executable behavior in the artifact.

Install only if you are comfortable letting the gws CLI access Gmail and GCP Pub/Sub for this workflow. Be careful with --output-dir because it stores email contents locally, and use --cleanup only when you want the watch session's created Pub/Sub resources removed on exit.

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
  • 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 (1)

Missing User Warnings

Low
Confidence
86% confidence
Finding
This markdown file describes a `--cleanup` flag that will delete created Pub/Sub resources on exit, which can affect system state. Although the flag name hints at cleanup, the description does not explicitly warn that resources will be deleted, so users may miss the destructive implication.

Static analysis

No suspicious patterns detected.