Tp2
- Category
- MCP Tool Poisoning
- Confidence
- 85% confidence
- Finding
Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
- Content
Security audit
Security checks for vulnerabilities and agentic risk
This skill appears to be a Chinese-language Amazon listing diagnostic helper with no hidden network, persistence, credential, or destructive behavior.
Install only if you are comfortable using a Chinese-language, heuristic Amazon listing advisor. Avoid pasting confidential launch plans, unreleased product details, or personal data, and treat the scoring as guidance rather than a guaranteed prediction of Amazon Rufus visibility.
Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
The skill promises a 5-dimensional diagnostic grounded in an industry knowledge base and Rufus visibility analysis, but the reported implementation behavior is only shallow heuristic scoring across fewer dimensions. This can mislead users into making business decisions based on overstated analytical capability, creating integrity and trust risks even if it is not direct code-execution abuse.
The manifest sets language: zh-CN, which is a natural-language locale constraint. The document does not indicate that users may choose another language or that the restriction is justified as a region-specific requirement, so it reads as a forced language policy.
The trigger phrase set is broad enough to match common Amazon/e-commerce help requests, which can cause the skill to activate outside a clearly consented or narrow diagnostic context. This is primarily a scope and routing risk: users may be steered into this skill unintentionally, leading to confusing behavior, unwanted processing of listing content, or overshadowing more appropriate general-assistance skills.
This markdown file is written entirely in Chinese and provides no indication that users may choose another language or that the content is intentionally limited to a Chinese-speaking audience. Under the policy rule for natural-language violations, forcing a specific language without opt-in is a locale/language policy issue.
The report formatter uses fixed Chinese labels and headings such as '标题', '五点描述', and 'Listing诊断报告', while the tool also accepts a '--lang' parameter. This creates a language-policy issue because users are not offered an actual language choice for output and the locale is effectively forced regardless of input.
The skill description is written as an instruction-facing README in Chinese and presents the workflow entirely in Chinese, including the required input format. There is no indication that users may choose another language or that the language restriction is necessary for a region-specific compliance purpose.
The skill invites users to paste product titles, bullets, and related business content but provides no warning about submitting confidential, embargoed, or otherwise sensitive commercial information. In a business-content optimization tool, this omission increases the chance of unintentional disclosure of proprietary marketing plans or unreleased product details.
The manifest provides the user-facing description only under the zh-CN locale, with no indication that other languages are supported or that use of Chinese is an intentional, documented locale restriction. This can amount to a language policy issue if the skill is surfaced to users without explicit language opt-in.