ClawHub
Back to skill

Security audit

CartPilot

Security checks for vulnerabilities and agentic risk

Overview

CartPilot is a disclosed shopping checkout-planning skill with a clear boundary against logging in, claiming coupons, placing orders, or payment.

Install this as an advisory checkout-planning skill. Users should provide only the cart, coupon, fee, ETA, and screenshot details they are comfortable sharing, and should keep all login, coupon-claiming, order submission, and payment steps under their own control.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The declared description says the skill optimizes shopping checkout paths across Chinese commerce platforms. The supplied code does not implement any checkout, coupon, order-splitting, platform comparison, or delivery optimization logic. Instead, it is a release automation script whose purpose is to validate and publish the skill/package metadata to Clawhub. This is a materially different primary purpose and includes undeclared external publishing behavior, so it is a clear mismatch.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The line explicitly says the release adds an "English README" and related publishing materials. This indicates a language choice in natural-language content without any note that users can choose another language or that English-only output is required for a justified region-specific purpose.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The description explicitly scopes the skill to 'mainland China shopping and local-delivery scenarios,' which imposes a locale constraint in the natural-language skill definition. The file does not offer a user choice of locale or clearly document this as an opt-in or compliance-bound regional limitation.

Content

No source excerpt is available for this finding.