同款服装多场景Lookbook｜AI-HIVE

• Outcome: Pass
• Latest audit: Aug 25, 2026 · 4:09 AM UTC
• Version: 1.0.0

Overview

The skill is an AI-HIVE lookbook image workflow that uses user-provided prompts, assets, and an API key; the main risks are expected external API use and billing, not hidden malicious behavior.

Install only if you are comfortable sending prompts and selected reference media to AI-HIVE and with possible paid generation calls after confirmation. Use environment variables or the init command carefully for the API key, avoid passing private files as reference images, and review model, routing, batch size, and price snapshot before submitting generation jobs.

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 (7)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 88% confidence
• Finding: The skill exposes and encourages use of capabilities including environment access, file read/write, network access, and shell execution, yet it declares no permissions. This creates a transparency and control gap: a user or host may invoke the skill without understanding that it can access local data, transmit content externally, or run commands that could affect the system.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 92% confidence
• Finding: The documented purpose is narrowly framed as lookbook generation, but the described behavior extends into broader chat, video generation, account and wallet access, model enumeration, API key setup/storage, and unrelated project brief generation. This mismatch is dangerous because it can hide unexpected data access, billing actions, credential handling, and broader external interactions behind a seemingly limited-use skill.

Description-Behavior Mismatch

Medium

• Confidence: 94% confidence
• Finding: The file is presented as a lookbook image-generation skill, but it actually ships a broad general-purpose client with chat, video, model enumeration, upload, and account-related functionality. This scope mismatch increases attack surface and makes it easier for a caller to invoke unintended capabilities that are unrelated to the declared skill purpose, violating least privilege and making review and policy enforcement harder.

Context-Inappropriate Capability

Medium

• Confidence: 91% confidence
• Finding: Exposing a user-info endpoint that returns account information and wallet balance is not necessary for generating lookbook images. In a skill context, this can leak sensitive account metadata to users or upstream orchestration layers and enables unnecessary data access beyond the advertised function.

Context-Inappropriate Capability

Medium

• Confidence: 89% confidence
• Finding: The general text-chat capability is outside the declared lookbook image-generation use case and gives the skill a broader execution surface than users and reviewers would expect. That can enable prompt injection, policy bypass attempts, or unreviewed content flows through a capability that should not be present in a narrowly scoped media-generation skill.

Context-Inappropriate Capability

Medium

• Confidence: 90% confidence
• Finding: Full video-generation support materially expands capability beyond the stated lookbook image skill. This creates unnecessary risk by allowing higher-cost, broader-content generation paths and media handling flows that are not required for the advertised task, increasing exposure to abuse and unintended policy violations.

Intent-Code Divergence

Medium

• Confidence: 83% confidence
• Finding: The top-level documentation explicitly describes the module as a general-purpose AI capability tool, which contradicts the skill's declared lookbook-only purpose. This discrepancy is a security-relevant indicator because it obscures actual capability during review and can mislead operators about what the skill can do.

Static analysis

No suspicious patterns detected.