Back to skill

Security audit

Product Photo Kit by VisualGPT

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates product-photo-kit benign with high confidence: the scanners found no AIG issues and one medium-severity warning about sharing images and prompts with an external provider.

The verdict is benign, not risk-free. Before using the optional VisualGPT workflow, consider whether you are authorized to share the images and briefs with that provider, and avoid submitting confidential or sensitive material unless you accept its data-handling terms. The three bundled WebP images were outside the static analyzers’ coverage, so evidence about their contents is incomplete.

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
Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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 (1)

The upload workflow lacks a third-party data-sharing warning.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The workflow instructs the agent to upload the user's source photo and submit the shot prompt to VisualGPT without warning that this shares their content with an external provider. The detail-page and image-editor workflows likewise instruct uploads, but the guide never cautions against sending confidential product imagery or sensitive information.

Content

No source excerpt is available for this finding.