ClawHub
Back to skill

Security audit

Seedance AI视频导演｜AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed AI-HIVE video production helper with user-directed API calls, local media processing, and credential setup that fit its stated purpose.

Install only if you are comfortable giving the skill an AI-HIVE API key, uploading user-selected media to AI-HIVE, and running ffmpeg locally. Review prompts, media rights, routing mode, and expected cost before allowing generation tasks, and avoid storing real API keys in logs or shared project files.

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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill advertises executable workflows that use environment variables, shell commands, file I/O, network access, and local media processing, yet the file declares no explicit permissions boundary. That mismatch can lead users or orchestrators to invoke a skill with broader capabilities than expected, increasing the risk of unauthorized API use, local file access, or command execution in environments that rely on declared permissions for containment.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding
The documented purpose is a narrowly scoped video-director workflow, but the described and detected behavior extends into general AI-HIVE chat, image generation, balance/user-info access, model enumeration, generic uploads, and broad ffmpeg editing. This over-breadth weakens least-privilege expectations and creates room for unintended data access, surprise billing, privacy exposure, or use of the skill as a gateway to unrelated capabilities not justified by the stated purpose.

Static analysis

No suspicious patterns detected.