ClawHub
Back to skill

Security audit

房产样板间漫游视频｜AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed AI-HIVE real-estate video workflow helper with expected API, media upload/download, local editing, and credential setup behavior.

Install only if you are comfortable using an AI-HIVE API key and potentially billable generation calls. Review prompts, routing, model parameters, output paths, and uploaded media before running commands; use only authorized property/media assets, keep API keys out of prompts/logs/repos, and avoid custom AI_HIVE_BASE_URL values unless you explicitly trust the endpoint.

SkillSpector

Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill advertises executable workflows that use environment variables, filesystem access, network calls, and shell-based tooling, yet it declares no permissions or equivalent capability boundary. This creates a confused-deputy risk where a user or host may invoke code with broader privileges than reviewers expect, increasing the chance of secret exposure, arbitrary file access, or unintended external requests.

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding
The documented purpose is narrowly framed as a real-estate walkthrough production helper, but the described scripts and capabilities enable broader generic AI generation, media upload/download, and local ffmpeg manipulation. This mismatch undermines user consent and security review because a seemingly specialized skill can be used as a general-purpose content and file-processing tool with networked side effects and potential billing impact.

Natural-Language Policy Violations

Medium
Confidence
91% confidence
Finding
The default prompt hard-codes a production-ready Chinese workflow and output format, which overrides user language preference without explicit opt-in. This can cause unintended behavior, reduce user control, and in multilingual contexts may be used to steer outputs away from what the user requested, though it is not a severe security issue by itself.

Static analysis

No suspicious patterns detected.