ClawHub
Back to skill

Security audit

Insurance Anti Fraud

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed, advisory-only insurance anti-fraud reference skill with sensitive subject matter but no bundled execution, persistence, credential handling, or hidden data flow.

Use this as an internal reference only. Before relying on its China-specific regulatory statements, scoring thresholds, or investigation workflows, verify them with current official sources and qualified legal/compliance reviewers, especially before denying claims or reporting cases externally.

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)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The description states the skill is built for the Chinese insurance market and includes Chinese-language policy framing and terminology, but it does not explicitly offer a language or locale choice to users. Because natural-language policy violations apply to all file types, this can be a locale-policy issue when the skill appears to force a specific regional context without opt-in.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger list includes very broad terms such as anti-fraud, risk control, claim review, and compliance-related Chinese phrases that could match many ordinary insurance workflows. This can cause unintended invocation of a powerful, fraud-focused skill in benign contexts, leading to overcollection prompts, inappropriate investigative framing, or misleading high-risk guidance being surfaced when the user did not request fraud analysis.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.