ClawHub
Back to skill

Security audit

gold-investment

Security checks for vulnerabilities and agentic risk

Overview

This skill is a focused gold-market analysis helper that uses Tavily search as disclosed, with no evidence of persistence, hidden execution, data destruction, or unrelated access.

Install only if you are comfortable providing a Tavily API key and having gold-market search queries sent to Tavily. Avoid including personal account details, confidential portfolio information, or regulated client data in prompts used with this skill.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (9)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill declares use of an environment variable and an external search API, but it does not define an explicit tool scope such as allowed-tools or permissions. This creates an authorization ambiguity where the runtime may permit broader network or execution behavior than reviewers and users can easily verify, increasing the chance of unintended data access or outbound requests.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger description is broad enough to activate on many ordinary gold-related questions, which can cause the skill to run unexpectedly and send user content to external services. Over-broad invocation increases the attack surface for unnecessary data exposure and unwanted tool use, especially in conversational contexts where users may not expect a networked finance-analysis skill to engage.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The documentation states that user queries are sent to Tavily for external search, but it provides no explicit user-facing privacy notice or guidance about what data may leave the system. If users include sensitive financial context, account details, or personal information in their prompt, that information could be transmitted to a third party without informed consent.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The example invocation phrases are very general and reinforce broad activation behavior without clear constraints. This makes accidental triggering more likely and can lead to external query transmission or analysis on loosely related user prompts without clear user intent.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The script sends both the user's search query and the Tavily API key to an external third-party service. While the API key must be transmitted for the request to work, the script provides no explicit disclosure, consent step, or data-handling warning, which creates a privacy and data-governance risk if users submit sensitive investment queries or internal prompts.

External Transmission

Medium
Category
Data Exfiltration
Content
include_raw_content: false,
  };

  const resp = await fetch("https://api.tavily.com/search", {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
Confidence
90% confidence
Finding
This finding is the same underlying issue as the fetch call: the script contacts api.tavily.com and sends request contents off-box. The skill context makes this somewhat less suspicious because web search is its stated purpose, but it remains a genuine privacy/compliance concern rather than a false positive.

External Transmission

Medium
Category
Data Exfiltration
Content
include_raw_content: false,
  };

  const resp = await fetch("https://api.tavily.com/search", {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
Confidence
90% confidence
Finding
This finding is the same underlying issue as the fetch call: the script contacts api.tavily.com and sends request contents off-box. The skill context makes this somewhat less suspicious because web search is its stated purpose, but it remains a genuine privacy/compliance concern rather than a false positive.

Natural-Language Policy Violations

Low
Confidence
95% confidence
Finding
This markdown file presents all guidance in Chinese and does not indicate that the user can choose another language or that the content is intentionally limited to a Chinese-speaking audience. Under the language/locale policy, forcing a specific language without user opt-in can be a policy violation.

Natural-Language Policy Violations

Low
Confidence
79% confidence
Finding
The natural-language description uses Chinese ('获取黄金相关信息') and the built-in default queries are exclusively Chinese, which indicates a language-specific behavior. The file does not state that the tool is region-specific or provide any opt-in or language choice.

Static analysis

No suspicious patterns detected.