Security audit

logo-designer-generator

Security checks for vulnerabilities and agentic risk

Overview

This is a user-directed remote image-generation skill with some labeling and privacy-disclosure gaps, but no hidden code, persistence, or destructive behavior in the inspected artifact.

Before installing, treat this as a hosted image/carousel generator rather than only a logo tool. Do not submit confidential brand assets, unreleased business material, or sensitive images unless you are comfortable sending them to the DeepNLP/Craftsman service under its terms. Review the npx onekey CLI separately before using that path.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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)

Description-Behavior Mismatch

Medium

Confidence: 93% confidence
Finding: The skill is presented as a logo designer, but the documentation exposes a broader remote image/post/carousel generation capability with multiple templates and content flows. This mismatch can mislead users and reviewers about what data is sent, what outputs are produced, and what external functionality is actually being invoked, increasing the chance of unsafe use and bypass of capability-based review.

Description-Behavior Mismatch

Low

Confidence: 84% confidence
Finding: The examples center on educational AI posts and slide-like carousels rather than logo generation, contradicting the advertised purpose of the skill. While not directly exploitable code execution, this deceptive or sloppy framing weakens informed consent and can cause users to submit broader content and assets to an external service than they intended.

Missing User Warnings

Medium

Confidence: 97% confidence
Finding: The documentation instructs users to send prompts, uploaded images, session metadata, and design configuration to a third-party remote API, but provides no privacy, retention, or transmission warning. In this context, users may upload proprietary images, branding material, or sensitive business content, so undisclosed external transmission creates a real confidentiality and compliance risk.

Static analysis

No suspicious patterns detected.