ClawHub
Back to skill

Security audit

Volcengine Image Generation

Security checks for vulnerabilities and agentic risk

Overview

The skill appears to generate images as advertised, but it needs review because local reference images are sent to a remote API and can be exposed in error output without a clear warning.

Review before installing. Use a scoped API key, verify the endpoint, avoid private or regulated reference images, and disable or redirect automatic downloads if Desktop writes are not desired. Be aware that failed requests may print prompts and base64-encoded local image data into logs or chat output.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill instructs users to run a bundled Python script that reads environment variables, accesses local files, performs outbound network requests, writes downloaded content to disk, and invokes shell/OS integration, yet it declares no permissions or equivalent safety disclosure. This mismatch can cause users or orchestration systems to grant or execute broader capabilities than expected without informed consent, especially because the workflow includes automatic downloads and opening Finder.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly supports passing local image paths and states they are automatically converted to base64 data URLs and sent to the remote image-generation API, but it does not provide a clear privacy warning at the point of use. Users may assume a local preprocessing step only, while actually transmitting potentially sensitive images off-device to a third party.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The script accepts local image paths, base64-encodes the file contents, and transmits them to a remote image-generation API without any explicit warning, confirmation, or allowlist control. In a skill designed to handle image upload workflows, this increases the chance that users unintentionally exfiltrate sensitive local files or images to an external service.

Static analysis

No suspicious patterns detected.