ClawHub
Back to skill

Security audit

TikTok-社媒运营专家 | 青虎AI

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed TikTok trend-research helper that uses Qinghu's authenticated API with user confirmation before calls, with no hidden execution or persistence found.

Before installing, users should understand that this skill uses Qinghu's external API, requires a Qinghu token, may consume Qinghu credits, and may export larger result sets to local files. Confirm calls only when the request really needs TikTok data research.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The activation description says users asking about TikTok hot topics, topic ideas, content trends, competitors, or content calendars 'must' trigger the skill, which is very broad and mandatory. That can cause over-triggering on generic marketing or planning requests and route users into paid external data collection flows unnecessarily, increasing the chance of unwanted tool use, cost, and disclosure of user intent to third-party services.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger examples are short and ambiguous, with no negative examples or boundaries to distinguish this skill from general brainstorming, calendar planning, or broader social-media advice. In context, that is more dangerous because the skill is user-invocable and explicitly instructs use of authenticated third-party APIs with possible point costs, so ambiguous triggering can lead to unnecessary external calls and charges.

Static analysis

No suspicious patterns detected.