ClawHub
Back to skill

Security audit

Character Consistency

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent image-generation guide for keeping a subject consistent across generated images, with a privacy caution around reference photos.

Before using this skill, treat reference images as data you are sending to the image-generation provider. Use only images you have rights and consent to process, avoid confidential or sensitive photos unless appropriate, and prefer controlled-access links over public URLs when possible.

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

Medium
Confidence
91% confidence
Finding
The examples explicitly encourage supplying reference images via URL, data URI, or base64, but do not warn that those images may contain personal, proprietary, or biometric data and will be transmitted to a third-party image service. In a skill centered on identity consistency across images, users are especially likely to upload sensitive portraits or product photos, so omission of privacy guidance can lead to unintended data disclosure or policy noncompliance.

Static analysis

No suspicious patterns detected.