ClawHub
Back to skill

Security audit

TikTok-单品分析师 | 青虎AI

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent TikTok product-analysis helper that uses a disclosed Qinghu API token and metered API calls, with some privacy and file-export points users should understand.

Install only if you intend to use Qinghu for TikTok Shop product analysis. Expect it to ask for or use a Qinghu API token, make metered API calls after confirmation, and create exported files for larger result sets; avoid installing it in environments where those token variables should not be available to analysis workflows.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger phrases are broad, everyday requests like 'analyze this TikTok product' or 'how is this product selling', which can cause the skill to activate in situations where the user did not intend to authorize external data access or paid tool usage. In this skill, over-triggering is more dangerous because the workflow also instructs collecting API credentials and making metered calls, so accidental invocation can lead to privacy exposure, unnecessary cost, or confusing context switching.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs automatic export of datasets to files and to provide links without first asking the user, which can create files containing potentially sensitive commercial data or derived analysis artifacts the user did not explicitly request. This is riskier in context because the export is mandatory once record counts reach a threshold, and file creation/link sharing expands the data surface beyond the immediate chat response.

Missing User Warnings

Low
Confidence
88% confidence
Finding
The skill explicitly directs reading API tokens from environment variables without any privacy warning, scope restriction, or instruction to prefer user-supplied secrets through the platform's secure mechanism. In an agent setting, normalizing secret access from environment variables can lead to unnecessary secret exposure or use of credentials the user did not realize would be consumed for this task.

Static analysis

No suspicious patterns detected.