ClawHub
Back to skill

Security audit

TikTok-蓝海爆品采集师 | 青虎AI

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent TikTok product research helper that uses a disclosed Qinghu API token and may export larger result sets to local files.

Before installing, make sure you are comfortable providing a Qinghu API token and with larger product-research outputs being written to local table files. Use a scoped token if available, and delete exported files when the sourcing research is no longer needed.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The skill explicitly instructs the agent to read API credentials from environment variables (`QINGHU_TOKEN` / `QHKIT_TOKEN`) even though the user-facing purpose is product research and sourcing. That creates an unnecessary secret-access capability: a compromised or overly broad skill could exfiltrate or misuse ambient credentials without a narrowly scoped, user-mediated auth flow.

Context-Inappropriate Capability

Medium
Confidence
88% confidence
Finding
The skill mandates automatic export of result sets to files and requires later reuse of local cached file paths. This expands the skill from remote data retrieval into local file creation and file-path handling, which can expose sensitive data to unintended persistence, cross-task access, or leakage through subsequent prompts and tools.

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The skill tells the agent to export large datasets to local files by default without presenting a nearby warning about storage, retention, or data sensitivity. In context, the exported material can include commercial research outputs and supplier data; silent persistence increases the risk of user surprise, unauthorized reuse, and accidental disclosure.

Static analysis

No suspicious patterns detected.