ClawHub
Back to skill

Security audit

互动分支短剧生成器｜AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed AI-HIVE short-drama workflow helper that uses user-provided API keys and media, with no hidden or destructive behavior found.

Install only if you are comfortable using AI-HIVE for media generation. Treat generation as potentially billable, keep the API key private, and only pass local images, videos, audio, or output paths that you intentionally want the scripts to read, upload, edit, or write.

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
83% confidence
Finding
The skill documents and demonstrates code paths that can access environment variables, local files, the network, and shell tools, yet it declares no permissions or capability boundaries. That creates a trust and containment gap: a caller or reviewer cannot tell what the skill may access, and sensitive data such as API keys or local media files could be exposed or modified through undocumented behavior.

Tp4

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding
The documented purpose is a narrow interactive short-drama generator, but the behavior reportedly includes generic AI chat, standalone ffmpeg/media manipulation, API-key setup and local storage, and generic upload/poll/download utilities. This mismatch is dangerous because it hides materially broader capabilities than users expect, increasing the chance of unauthorized file processing, credential handling, network exfiltration, or misuse under the cover of a benign-seeming skill description.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The manifest sets allow_implicit_invocation to true without any visible trigger constraints, allowing the skill to be auto-selected outside clearly bounded user intent. Because this skill can generate runnable workflows and AI-HIVE commands, unintended invocation could cause the agent to steer users into tool-backed content generation or policy-sensitive workflows they did not explicitly request.

Natural-Language Policy Violations

Medium
Confidence
85% confidence
Finding
The default prompt hard-codes production-ready Chinese output regardless of the user's preferred language. While not directly a code-execution issue, this can override user intent, reduce transparency, and make outputs harder for some users or reviewers to validate, which is riskier in a skill that emits runnable commands and workflow instructions.

Static analysis

No suspicious patterns detected.