Undeclared Tool Scope
Medium
- Category
- MCP Least Privilege
- Confidence
- 70% confidence
- Finding
Without declared permissions the skill's intent is opaque and cannot be validated.
- Content
Security audit
China E-Commerce Shopping Guide
Security checks for vulnerabilities and agentic risk
Overview
This is a static China e-commerce shopping guide with only minor wording and routing issues, not hidden or high-risk behavior.
Install this only if you want static guidance for Mainland China e-commerce platforms. Treat recommendations as general reference, verify current platform policies and seller details yourself, and be aware broad shopping queries may receive China-specific advice.
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
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
- 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
Findings (9)
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 92% confidence
- Finding
The trigger phrases are broad enough to match common shopping questions that many unrelated assistants or skills could receive, which can cause unintended invocation. That creates routing and prompt-surface risk, where users may be shifted into this skill without clear intent, potentially leading to irrelevant advice, locale-mismatched recommendations, or increased exposure to any downstream issues in the skill implementation.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 80% confidence
- Finding
The description states the skill is focused on recommending mainstream Chinese e-commerce platforms, but does not clearly frame this as a region-specific skill limitation or require user intent for China-market shopping. This can create a locale-policy issue if users are routed into a China-specific recommendation flow without an explicit opt-in or region check.
- Content
Intent-Code Divergence
Low
- Category
- Not specified by scanner
- Confidence
- 89% confidence
- Finding
The module docstring and manifest frame the skill as a China-focused shopping guide, and the implementation only loads local category/platform data for Chinese platforms. The help banner's phrase '全球在线购物指南' contradicts that narrower actual behavior rather than merely omitting detail.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 84% confidence
- Finding
This file is a code file, so SQP-3 applies to natural-language strings in the help output. The help text presents the experience only in Chinese and labels it 'global', but the tool behavior and branding are specific to Chinese e-commerce platforms, with no user choice of language or locale.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 88% confidence
- Finding
This JSON file contains natural-language category keywords in both Chinese and English across many entries, but provides no indication that the skill supports multiple languages by design or how language choice is determined. Under the language/locale policy rule, silently imposing or assuming a locale behavior without documented opt-in can be a policy concern.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 84% confidence
- Finding
This JSON consistently encodes user-facing platform descriptions such as strengths, weaknesses, notes, and policy fields in English, while the dataset is explicitly scoped to Mainland China and includes Chinese names. Under the natural-language policy rule, this can be a locale/language policy issue because the file provides no opt-in or documented justification for forcing a single language.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 80% confidence
- Finding
The guide is entirely framed around China-specific platforms and shopping events such as 618 and Double 11, presenting the content as a general platform guide rather than explicitly as a region-specific reference. Because the policy requires either user choice or clear justification for locale constraints, this broad presentation can be read as imposing a specific locale context without opt-in.
- Content
Description-Behavior Mismatch
Low
- Category
- Not specified by scanner
- Confidence
- 92% confidence
- Finding
The manifest says the skill is focused on recommending mainstream Chinese e-commerce platforms such as Taobao, Tmall, JD.com, and Pinduoduo. This reference file also provides platform-specific guidance for Douyin, and elsewhere mentions Vipshop and Xiaohongshu, which expands the scope beyond the described platform set.
- Content