ClawHub
Back to skill

Security audit

LiblibAI 哩布哩布 libtv 图片生成替代｜AI 图片生成与编辑

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent image-generation wrapper that uses an external AI Hive API, stores an API key locally, and uploads only user-selected image inputs without hidden destructive behavior.

Install only if you are comfortable sending prompts and any selected reference images to the AI Hive service and storing an AI Hive API key locally. Use only images you have rights to provide, and remove ~/.ai-hive/config.json if you no longer want the local key stored.

SkillSpector

Vulnerability Patterns
  • 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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill advertises executable capabilities including environment access, file read/write, network, and shell usage, yet declares no permissions. This weakens reviewability and consent because operators and users cannot accurately assess what the skill may access or modify before use. In this context, networked image-generation commands, local config storage, and shell invocation increase the chance of unintended data exposure or misuse of local resources.

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding
The documented purpose is narrowly framed as translating third-party workflow characteristics into observable visual tasks, but the skill also appears to support API-key onboarding, persistent local config storage, standalone uploads, arbitrary task lookup, and direct prompt submission to a fixed external service. This mismatch is dangerous because reviewers and users may invoke the skill under false assumptions, while it actually performs broader external-data transfer and account-related actions than advertised.

Vague Triggers

Medium
Confidence
79% confidence
Finding
The invocation text is broad enough to match generic image-generation and editing requests, not just the stated LiblibAI-migration use case. Overbroad triggering can cause the skill to handle unrelated prompts, leading to unnecessary external API use, unintended uploads, or policy bypass through use outside its reviewed context.

Static analysis

No suspicious patterns detected.