ClawHub
Back to skill

Security audit

胡开奕 · 养老与传承顾问技能 · 会谈纪要整理

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Chinese-language wealth-planning meeting-notes skill with sensitive-data handling risks, but no hidden code, persistence, network execution, or unrelated authority.

Install only if you intend to use it for Chinese-language养老与传承 client meeting-note workflows. Before pasting transcripts or chats, remove names, phone numbers, IDs, account numbers, addresses, medical record identifiers, and any third-party details that are not needed; confirm you have permission to process client or family information. Consider narrowing the generic triggers if your agent auto-invokes skills.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

This skill is explicitly designed to process highly sensitive personal and family data, including health status, finances, family relationships, insurance coverage, and potentially legal/planning details. Although it mentions脱敏 for external outputs, it does not provide an upfront privacy/data-handling warning or clear minimization guidance before users paste raw transcripts, chats, or notes, which increases the risk of unnecessary disclosure of regulated or confidential data into the agent workflow.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger list contains generic phrases such as “会谈纪要”, “整理纪要”, and “客户沟通纪要” that could match ordinary user requests unrelated to this specific skill. This can cause unintentional skill invocation, exposing sensitive meeting-note processing behavior in the wrong context and increasing the chance that private client information is routed into a wealth-management workflow unexpectedly.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

This markdown file contains user-facing natural-language content that assumes Chinese as the required language. Under the policy, forcing a specific language without user opt-in can be a locale-policy violation, and the file does not indicate that the skill is limited to Chinese-speaking users or offers alternatives.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.