ClawHub
Back to skill

Security audit

TikTok-视频复刻专家 | 青虎AI

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Qinghu/TikTok research skill that uses an external API after user authorization, with no hidden code, persistence, or destructive behavior found.

Install only if you want Qinghu-backed TikTok video research and are comfortable providing or exposing a Qinghu API token. Review tool calls before approving them, because calls may consume Qinghu credits and send request details to Qinghu.

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 (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill description says it 'must trigger' for broad user intents like cloning TikTok videos, script breakdowns, content templates, and reducing script originality cost. These requests overlap heavily with normal creative-assistance tasks, so mandatory invocation can hijack unrelated conversations and unnecessarily steer users into external-tool workflows that may incur cost or expose data. The broadness is more concerning here because the skill also pushes tool/API usage and external authorization flows, increasing the consequence of accidental activation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
81% confidence
Finding

The file consistently specifies the skill in Chinese and presents all instructions, outputs, and user-facing guidance only in Chinese, with no indication that users may choose another language. Under the stated policy, forcing a specific language without opt-in is a natural-language policy concern unless the locale constraint is explicitly justified.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The example trigger phrases include generic requests such as 'what kind of videos can sell products' and 'give me a script template,' which are not unique to this skill. If used for routing, these examples can cause the skill to activate during ordinary marketing or writing help, leading to overcollection of user intent, unnecessary external calls, and paid operations unrelated to the user's actual need.

Content

No source excerpt is available for this finding.