抖音电商视频生成与编辑

• Outcome: Pass
• Latest audit: Aug 17, 2026 · 7:14 AM UTC
• Version: 1.0.0

Overview

This skill is a coherent AI Hive video-generation helper, with expected remote uploads and API-key storage that users should understand before use.

Install only if you are comfortable sending selected product images, videos, audio, and prompts to AI Hive for processing. Prefer using an environment variable or scoped API key if you do not want a persistent key stored under ~/.ai-hive, and avoid uploading confidential customer or unreleased business media unless your organization permits that service.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• 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

Findings (8)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 82% confidence
• Finding: The skill advertises executable commands that can read environment variables, access files, invoke shell commands, and make network requests, but it does not declare corresponding permissions. This creates a transparency and least-privilege problem: users or hosting platforms cannot accurately assess what the skill may access before execution, increasing the risk of unintended data exposure or command execution.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 93% confidence
• Finding: The declared purpose is narrowly framed as Douyin e-commerce video generation/editing, but the documented behavior appears to include broader capabilities such as text chat, image generation, model enumeration, uploads, user-info access, and browser-assisted API key acquisition. This mismatch is dangerous because it can mislead users and reviewers about the actual attack surface, especially where account data, API credentials, or unrelated remote actions are involved.

Description-Behavior Mismatch

Medium

• Confidence: 84% confidence
• Finding: The file contains generic capabilities for user-info, model enumeration, text chat, and image generation in addition to the advertised Douyin video workflow. Even though the skill entrypoint currently uses the restricted parser, bundling broader functions increases the chance of accidental exposure, repurposing, or future invocation outside intended scope, which expands the attack surface and data accessible through the skill.

Context-Inappropriate Capability

Medium

• Confidence: 80% confidence
• Finding: The user-info endpoint retrieves account/profile and wallet balance information, which is unnecessary for Douyin e-commerce video generation/editing. If exposed through this skill or reused by another entrypoint, it could disclose sensitive account metadata unrelated to the user task, violating least privilege.

Context-Inappropriate Capability

Medium

• Confidence: 78% confidence
• Finding: The generic text-chat function is unrelated to the declared video-editing purpose and can send arbitrary prompts and attached media to the remote API. In a skill meant only for video generation/editing, this broadens functionality beyond user expectations and can be abused to exfiltrate task data or repurpose the skill as a general LLM conduit.

Context-Inappropriate Capability

Medium

• Confidence: 79% confidence
• Finding: Standalone image generation exceeds the declared purpose of a video-generation/editing skill. While not inherently malicious, retaining unrelated generation paths increases attack surface and may cause users to send prompts or reference media to remote services under a misleading scope assumption.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The upload flow transmits local media files to remote API/object storage endpoints without an explicit privacy warning at the point of upload. In this skill context, users may upload product, customer, or internal marketing assets; silent transmission to third-party infrastructure can lead to unintended disclosure of sensitive business media.

Missing User Warnings

Medium

• Confidence: 83% confidence
• Finding: The init flow persists the API key to a local config file without a strong sensitivity warning before storage. Although permissions are tightened to 0600, users may still unknowingly leave long-lived credentials on disk, increasing exposure through backups, endpoint compromise, or shared account misuse.

VirusTotal

65/65 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.