Back to skill

Security audit

Maicai

Security checks for vulnerabilities and agentic risk

Overview

This skill is a transparent China grocery price-comparison helper and does not show hidden execution, credential use, ordering, persistence, or data exfiltration behavior.

Install this if you want help comparing public grocery prices for mainland China platforms. Do not use it for private account pricing or checkout automation; the artifact itself says to stop before login, coupon claiming, payment, or ordering.

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
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (5)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill strongly mandates a Chinese-language persona and preferred phrasing without indicating that the assistant should adapt to the user's requested language. This can override user preference, reduce accessibility, and create a prompt-quality/safety issue where the agent follows skill style constraints over the user's explicit communication needs.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The default prompt begins with 'Use $maicai' and then specifies all instructions in English, while the display name and short description are in Chinese and the skill targets mainland China grocery platforms. This creates a natural-language locale inconsistency and effectively imposes a language choice without documenting user opt-in or offering alternatives.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The rules prescribe exact Chinese fallback phrases, which can override user language preference and cause the agent to respond in Chinese without opt-in. This is primarily a prompt-quality and user-control issue rather than a direct security exploit, but it can degrade trust, accessibility, and policy compliance in multilingual contexts.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The natural-language description forces a single language presentation but does not state that the skill is China- or Chinese-language-only, nor does it offer any user opt-in or language choice. Under the policy, a fixed language without documented justification or choice is a locale/language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

The reference is written as if the applicable platforms and pricing context are fixed, including Chinese platform names and a tie-breaker mentioning saving 'a few yuan'. This can be a language/locale policy concern because it assumes a specific regional market context without stating that the skill is China-specific or giving the user an opt-in choice.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.