Back to skill

Security audit

爆款广告变量拆解|AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed AI-HIVE advertising-analysis and media-generation helper, with expected API-key, upload, download, and ffmpeg behavior for that purpose.

Install only if you are comfortable sending selected media files and prompts to AI-HIVE, using an AI-HIVE API key, and potentially incurring generation charges after confirmation. Use authorized reference materials, avoid storing API keys in shared environments, and review generated parameters before submitting tasks.

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
87% confidence
Finding
The skill instructs users to run local scripts that use environment variables, file I/O, shellable tooling like ffmpeg, and network access, yet no permissions are declared. This weakens reviewability and informed consent because an agent or user may invoke capabilities broader than the metadata suggests, including uploading local media or using API credentials without clear gating.

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding
The documented purpose is a scoped ad-variable analysis workflow, but the embedded commands and described behavior enable broader generic chat, media generation, uploads, task polling/downloads, and local ffmpeg editing. This mismatch is dangerous because it can mask a much wider execution and data-handling surface than reviewers or users expect, increasing the chance of unauthorized uploads, excessive spend, or misuse of local files.

Static analysis

No suspicious patterns detected.