ClawHub
Back to skill

Security audit

Virtual Try-On

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed virtual try-on prompt workflow with consent safeguards and no bundled code, persistence, or hidden execution.

Before installing, make sure users understand this skill processes likeness-sensitive images through whatever image generator is already available, and only use it with photos and garment images they own or have permission to use.

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
  • 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
Findings (3)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% 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 · README.md (reported line 25)May include surrounding context.

md
## Rights

Use a photo you own or have permission to use. Do not run this on someone else's likeness without consent. Outputs are AI-generated — label them that way. Do not use results as fake ads or fake endorsements.

## Full version

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger description includes broad natural-language phrases such as 'see how this looks on me' and 'put this outfit on this person,' which can match ordinary conversation and cause the skill to activate unintentionally. In this skill, unintended invocation could route users into an image-manipulation workflow involving likeness and consent-sensitive content, increasing the chance of accidental processing of risky requests.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
94% confidence
Finding

The invocation guidance uses the phrase "try this outfit on this person," which is fairly general everyday language rather than a narrowly scoped command. The README does not provide explicit trigger constraints or negative examples to clarify when the skill should or should not activate.

Content

No source excerpt is available for this finding.