商品短视频脚本生成器｜AI-HIVE

• Outcome: Pass
• Latest audit: Aug 25, 2026 · 10:26 AM UTC
• Version: 1.0.0

Overview

This skill is a disclosed AI-HIVE video workflow helper that uses an API key, user-selected media, and output files, with no evidence of hidden exfiltration or destructive behavior.

Install only if you are comfortable giving the skill access to your AI-HIVE API key and uploading the media files you explicitly pass to it. Review prompts, routing, and pricing before running generation commands, and keep the stored API key file private.

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 (6)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 88% confidence
• Finding: The skill exposes code paths that use environment variables, file read/write, network access, and shell execution, but it declares no permissions or capability boundaries. That mismatch is dangerous because consumers and orchestrators cannot accurately assess or sandbox the skill, increasing the chance of unintended credential access, local file exposure, or execution of external commands during normal use.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 93% confidence
• Finding: The documented purpose is a narrowly scoped product video script generator, but the described behavior includes broader capabilities such as account info retrieval, wallet balance access, generic chat/image generation, model enumeration, and local credential setup/storage. This creates a trust-boundary violation: users may authorize or invoke the skill expecting limited storyboard generation while it can access sensitive account data and perform unrelated networked actions, which materially increases abuse and data-exposure risk.

Description-Behavior Mismatch

High

• Confidence: 95% confidence
• Finding: The file is presented as a specialized product video script generator, but the implementation is a general-purpose AI Hive client that supports chat, model enumeration, media upload, task polling, downloads, and account-related operations. This capability mismatch increases the chance that callers grant broader trust, permissions, and sensitive inputs than the advertised skill purpose requires, enabling misuse beyond the stated business function.

Context-Inappropriate Capability

Medium

• Confidence: 87% confidence
• Finding: The skill includes a user-info endpoint that retrieves account and wallet/balance data, which is not necessary for generating product video scripts. In a skill context, unnecessary access to billing or account data expands data exposure and can leak sensitive operational information to users or upstream orchestrators that only expected creative-generation behavior.

Context-Inappropriate Capability

Medium

• Confidence: 84% confidence
• Finding: The generic text chat function exceeds the skill's advertised purpose of generating product video scripts and can be repurposed for arbitrary model interaction. This broadens the attack surface and allows prompt routing, data submission, and model access that users may not realize they are enabling when invoking a narrowly branded skill.

Intent-Code Divergence

Medium

• Confidence: 90% confidence
• Finding: The top-level documentation describes a general AI capability tool rather than the advertised specialized script generator, creating a transparency and trust-boundary problem. Users, reviewers, and calling agents may make unsafe assumptions about allowed behavior, data handling, and privilege scope because the skill branding and code behavior do not align.

Static analysis

No suspicious patterns detected.