ClawHub
Back to skill

Security audit

Persona Customer Support

Security checks for vulnerabilities and agentic risk

Overview

This customer-support persona is a simple, disclosed Google Workspace workflow guide with no bundled code, hidden persistence, or deceptive behavior.

Install this only for a Google Workspace account, support inbox label, tracking sheet, Chat space, and calendar where the agent is allowed to handle customer-support work. Confirm messages, escalations, and calendar inserts before letting the agent perform them automatically.

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)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The skill description is broad enough to invite activation for general customer-support behavior without clearly limiting when it should be used or what authority boundaries apply. In a persona skill that can access Gmail, Sheets, Chat, and Calendar, ambiguous activation increases the chance of the agent being steered into processing sensitive communications or taking actions outside an intended support workflow.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The heading and summary present the skill as a general-purpose customer support agent without defining trigger conditions, permitted data sources, or action boundaries. Because the instructions enable triage, task creation, sheet logging, chat escalation, and calendar scheduling, unclear invocation constraints could let unrelated prompts cause real-world actions across multiple integrated services.

Static analysis

No suspicious patterns detected.