ClawHub
Back to skill

Security audit

gold-price-query

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple gold-price lookup that makes a disclosed third-party API request and does not show hidden persistence, credential access, or destructive behavior.

Install only if you are comfortable with the skill contacting jijinhao.com for each query and sending normal request metadata, including browser-like headers. It appears purpose-aligned, but a tighter version would declare network permissions explicitly, minimize headers, and add a request timeout.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding

The skill documentation indicates network access is required, but the manifest does not declare any explicit tool scope such as permissions or allowed-tools. This creates a least-privilege and transparency issue: reviewers and runtime policy systems cannot clearly constrain or validate the network capability the skill depends on, which can enable unintended outbound requests or make future abuse harder to detect.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · gold_price.py (reported line 3)May include surrounding context.

python
"""
黄金价格查询工具
查询接口: https://api.jijinhao.com/quoteCenter/realTime.htm
"""

import asyncio

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

The skill makes a live request to a third-party service, which causes external transmission of metadata such as IP address, User-Agent, Referer, and language preferences. In this skill's context, the transmitted payload is limited and does not include obvious secrets, so the main risk is privacy leakage and unreviewed dependency on an external endpoint rather than direct compromise.

Content

Scanner excerpt · gold_price.py (reported line 28)May include surrounding context.

python
# 根据货币类型选择对应的接口
        if currency.upper() == "CNY":
            url = f"https://api.jijinhao.com/quoteCenter/realTime.htm?codes=JO_92233&isCalc=true&_={timestamp}"
        else:
            url = f"https://api.jijinhao.com/quoteCenter/realTime.htm?codes=JO_92233&_={timestamp}"

External Transmission

Medium
Category
Data Exfiltration
Confidence
86% confidence
Finding

This request includes an unusually complete set of browser-mimicking headers, including Referer, User-Agent, and language/platform hints, which increases unnecessary metadata disclosure to the external service. While not overtly malicious, this makes the external transmission more privacy-invasive and may indicate intentional evasion of normal API access patterns.

Content

Scanner excerpt · gold_price.py (reported line 30)May include surrounding context.

python
if currency.upper() == "CNY":
            url = f"https://api.jijinhao.com/quoteCenter/realTime.htm?codes=JO_92233&isCalc=true&_={timestamp}"
        else:
            url = f"https://api.jijinhao.com/quoteCenter/realTime.htm?codes=JO_92233&_={timestamp}"
        
        # 添加完整的请求头，模拟浏览器行为
        headers = {

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The request sets accept-language to zh-CN,zh;q=0.9, which forces a specific locale preference in outbound requests. This is a natural-language/locale policy concern because the skill does not offer the user any language choice or explain why Chinese locale is required.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.