ClawHub
Back to skill

Security audit

E-commerce Scraper

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed ToolRouter-based e-commerce scraping helper, with expected third-party data sharing and minor setup/storage cautions.

Install only if you are comfortable sending target URLs, search queries, and scraped content to ToolRouter. Avoid internal, login-protected, regulated, or confidential pages without approval, consider pinning or reviewing the MCP package before use, and set include_images:false when you do not want images saved locally.

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 Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

The skill instructs users to connect to an external MCP endpoint at https://api.toolrouter.com/mcp, which means scraped URLs, queries, and returned content are transmitted to a third-party service. In this skill context, external transmission is expected, but it is still security-relevant because users may send proprietary competitor research, sensitive URLs, or other regulated content off-platform.

Content

Scanner excerpt · SKILL.md (reported line 13)May include surrounding context.

md
## Installation

Install with `pnpm dlx skills add Humanleap/agent-skills --skill ecommerce-scraper`. Connect the client's remote MCP server to `https://api.toolrouter.com/mcp` (Claude: Customize → Connectors → Add custom connector; Claude Code: `claude mcp add toolrouter -- npx -y toolrouter-mcp`). Setup: https://toolrouter.com/docs/markdown/quickstart.

## Hard Rules

Rp1

Medium
Category
MCP Rug Pull
Confidence
89% confidence
Finding

The installation instructions invoke npx -y toolrouter-mcp without pinning a specific package version. This creates a supply-chain risk because future package updates or a compromised published version could be executed by users automatically when they connect the MCP server.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill is presented primarily as a product and marketplace scraper, but the workflow explicitly expands to a general web-scraper/scrape_page capability for arbitrary pages. That scope expansion matters because it can cause users or downstream agents to send broader website content than expected to the external MCP service, increasing data exposure and policy/compliance risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The documentation notes only later that scrape_product saves images to the user's file space by default, but this behavior is not surfaced prominently in the skill description or upfront warnings. Hidden persistence of downloaded content can lead to unexpected storage of copyrighted, sensitive, or unnecessary files and increase data retention risk.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

L03 describes scraping product pages and searching marketplace listings for research and pricing checks, but does not mention any dedicated snapshot operation. L42 and L53 introduce watchlist_snapshot for tracking prices over time, which is an additional monitoring-oriented capability beyond the manifest's narrowly described scrape/search behavior.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.