ClawHub
Back to skill

Security audit

TikTok 选品、趋势、竞品与内容洞察【Gecho 官方】

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed TikTok insights integration that routes user-requested research jobs through Gecho and does not show hidden or destructive behavior.

Install only if you are comfortable using Gecho's Chrome extension and MCP bridge with a logged-in TikTok browser session. Review the Gecho extension/package provenance and remember this skill starts asynchronous research jobs and may save results to a directory, but it does not claim to post content or collect passwords.

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)

Natural-Language Policy Violations

High
Category
Not specified by scanner
Confidence
95% confidence
Finding

L003 的描述全文为中文，且整份技能文件均以中文固定说明用户交互与响应模板，没有看到向用户提供语言/地区选择或显式 opt-in 的机制。根据规则，强制单一语言而未提供选择属于语言/locale 政策问题。

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The example trigger phrases are broad enough that normal conversational requests like '分析 TikTok 上的 xxx' could cause the agent to invoke the Gecho workflow without a clear user intent to run an external tool. In this skill, tool use depends on a live logged-in browser session and third-party MCP service, so overbroad triggers increase the chance of unintended external actions, data access, or confusing task creation.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The execution-condition text says to use the skill whenever users ask about product analysis, trends, competitors, market opportunities, content strategy, or status checks, but it lacks tight boundaries distinguishing advice from actual tool execution. Because this skill can trigger asynchronous jobs through a third-party bridge tied to an authenticated TikTok session, ambiguous routing can lead to unintended tool calls and exposure of session-derived data.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.