ClawHub
Back to skill

Security audit

Fraud Prevention Guide

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown fraud-prevention guide with purpose-aligned but privacy-sensitive recommendations, and it does not install code or access systems.

Before installing or using this skill, treat it as policy-design guidance rather than a turnkey automation. If you implement its recommendations, add legal/privacy review, disclose tracking practices, minimize behavioral and fingerprinting data, set retention limits, restrict access, and require human review or rollback paths for high-impact decline/block rules.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The guide explicitly recommends collecting broad device, network, behavioral, and account telemetry, including browser fingerprinting, mouse movement patterns, geolocation, and transaction metadata, but provides no constraints on lawful basis, notice, minimization, retention, or access controls. In a fraud-prevention context this can lead implementers to deploy invasive tracking and store sensitive behavioral data in ways that create privacy, compliance, and secondary-security risks if overcollected or later breached.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 97)May include surrounding context.

md
| Score Range | Action | Processing |
|---|---|---|
| 0–15 | Auto-approve | Process immediately, no additional friction |
| 16–30 | Low-risk review | Approve with enhanced monitoring, flag for batch review |
| 31–50 | Manual review | Hold order for analyst review within 2 hours |
| 51–70 | Enhanced verification | Require additional authentication (3DS challenge, email/phone verify) |

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · references/risk-scoring-reference.md (reported line 63)May include surrounding context.

md
| Score Range | Action | Processing |
|---|---|---|
| 0–15 | Auto-approve | Process immediately, no additional friction |
| 16–30 | Low-risk review | Approve with enhanced monitoring, flag for batch review |
| 31–50 | Manual review | Hold order for analyst review within 2 hours |
| 51–70 | Enhanced verification | Require additional authentication (3DS challenge, email/phone verify) |