抖音电商图片生成与编辑

• Outcome: Pass
• Latest audit: Aug 17, 2026 · 5:49 AM UTC
• Version: 1.0.0

Overview

This skill is primarily a disclosed AI Hive image-generation tool, with some broader helper code and credential storage that users should understand before installing.

Install only if you are comfortable uploading selected product/reference images and prompts to AI Hive and storing an AI Hive API key locally. In stricter environments, ask the publisher for a trimmed image-only package with pinned dependencies and explicit permission metadata.

SkillSpector

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration

Findings (5)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 88% confidence
• Finding: The skill invokes local scripts with shell commands, performs package installation, and appears to rely on environment variables, filesystem access, and network connectivity, yet it declares no permissions. This creates a transparency and sandboxing problem: a caller may trust the skill as low-risk while it can read/write local data and make outbound requests.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence
• Finding: The documented purpose is narrowly framed as Douyin e-commerce image generation/editing, but the underlying behavior reportedly supports broader AI Hive operations such as text chat, video generation, model enumeration, account/wallet queries, and generic media upload. This mismatch is dangerous because it conceals materially broader capabilities that could be abused for data exfiltration, unintended API actions, or policy bypass under the guise of a narrow image tool.

Description-Behavior Mismatch

Medium

• Confidence: 90% confidence
• Finding: The file advertises and implements a general-purpose AI Hive client with text chat, image generation, video generation, model listing, uploads, and task polling, while the skill manifest presents it as a Douyin e-commerce image-generation/editing skill. This capability mismatch increases the attack surface and violates least functionality: a caller or wrapper expecting only image-editing behavior may unknowingly expose broader model access and non-image workflows.

Context-Inappropriate Capability

Medium

• Confidence: 92% confidence
• Finding: The code contains a full video-generation API path even though this skill is scoped to e-commerce image creation/editing. Extra media-generation functionality broadens what an invoking agent can do with the same credentials and can be abused for unauthorized or policy-bypassing content generation outside the stated skill purpose.

Context-Inappropriate Capability

Low

• Confidence: 86% confidence
• Finding: The presence of a generic text chat capability is outside the stated purpose of an e-commerce image skill. While not inherently dangerous by itself, it creates hidden functionality that can be misused by an agent or user to send arbitrary prompts and attachments to remote models using the configured API key.

VirusTotal

65/65 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.