ClawHub
Back to skill

Security audit

潮汐调研框架

Security checks for vulnerabilities and agentic risk

Overview

This is a text-only research workflow skill with no code execution, persistence, credential use, or hidden data handling.

Install this if you want a structured deep-research workflow. Be aware it may activate on broad research-like prompts and may produce Chinese section labels unless you explicitly ask for another language.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger list is broad enough to match common, non-specialized requests such as '分析一下', 'research report', or '帮我研究一下', which can cause the skill to activate when the user did not actually want this heavyweight workflow. Unintended activation can override normal interaction patterns, increase tool usage, and push the agent into unnecessary web research or structured output modes that may expose more data or produce confusing behavior.

Natural-Language Policy Violations

Medium
Confidence
89% confidence
Finding
The mandated output labels and section names are in Chinese without a documented language-selection rule, which can cause the skill to ignore the user's preferred language and reduce usability or clarity. While this is not a direct security exploit, it can lead to miscommunication in research outputs, especially when the user expects English or another language for downstream review or automated processing.

Static analysis

No suspicious patterns detected.