Security audit

吊顶装修指南

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only Chinese home-renovation advice skill with some over-broad reference content, but no hidden execution, credential access, persistence, or deceptive behavior was found.

Install only if you want Chinese-language renovation guidance from this specific style of source. Treat the broader non-ceiling references as general renovation context, and verify any structural, exterior, HVAC, electrical, waterproofing, or property-rule advice with qualified local professionals before acting.

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

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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 (169)

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 99% confidence
Finding: The reference content is materially misaligned with the skill’s declared purpose of suspended-ceiling renovation guidance, focusing instead on wall-mounted air-conditioner installation. This kind of scope drift can cause the agent to retrieve and present irrelevant or misleading advice in a construction context, increasing the chance of poor recommendations and user harm from incorrect renovation decisions.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The file content is substantially about kitchen air-conditioning selection and installation rather than the declared ceiling-renovation guidance domain. This creates a scope-integrity problem: the agent may retrieve and present irrelevant advice, reducing trustworthiness and potentially leading users to make decisions outside the skill’s stated expertise.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 99% confidence
Finding: The file content is materially off-topic for a ceiling-decoration skill: it focuses on repurposing air-conditioner outdoor-unit spaces, including enclosure, demolition, storage, pet areas, and laundry placement. In a specialized装修指南 skill, this kind of domain drift can mislead users into following unrelated renovation advice, some of which touches building-envelope changes and equipment-access constraints without proper qualification.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 99% confidence
Finding: The referenced content is about balcony usage, leisure, office, and laundry layouts rather than the declared ceiling-renodeling guidance domain. This creates a skill-integrity problem: users relying on the manifest for ceiling-specific renovation advice may receive off-topic recommendations, which can reduce trust and lead to poor renovation decisions in a domain where technical constraints matter.
Content: No source excerpt is available for this finding.

Hidden Instructions

High

Category: Prompt Injection
Confidence: 60% confidence
Finding: Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.
Content: Scanner excerpt · references/ref-198.md (reported line 66)May include surrounding context.
md
——风格符合就行，呵呵，有人是经常要换得。过去直接用床单挂上就行的，看个人财力了 ‍‍ ✪公众号x ✪

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 99% confidence
Finding: The referenced content is materially misaligned with the skill's declared purpose: it discusses whole-home custom cabinetry, contract details, appliance coordination, and installation acceptance rather than ceiling-decoration guidance. This can cause the agent to retrieve and present irrelevant or incorrect advice in response to user requests, weakening reliability and potentially leading users to make bad renovation decisions based on the wrong domain knowledge.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The file is materially out of scope for a ceiling-decoration skill: it contains broad advice on wardrobes, cabinets, sofas, appliances, smart home, curtains, and soft furnishings rather than focused ceiling guidance. This can cause the agent to answer with irrelevant or misleading content, weakening safety and reliability boundaries and increasing the chance of inappropriate recommendations in contexts where users expect specialized吊顶 advice.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The instruction says to use the '西安装修课堂' tone specifically, which imposes a particular language/locale style on all responses. Under the policy, forcing a specific language or locale without user opt-in is a natural-language policy concern.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: This line directs the agent to respond in the '西安装修课堂' everyday style, with no indication that users may choose another language or locale. That creates a policy issue because the skill hard-codes a locale-specific communication style across interactions.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The entire file is written in Chinese and provides no indication that users may choose another language or locale. Under the policy criteria, content that enforces a specific language without user opt-in can be a natural-language policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: This markdown file is entirely written in Chinese and includes region-specific guidance such as '西安地区' and '发布地点: 陕西', but it does not indicate that the language/locale is optional, user-selected, or intentionally restricted to a justified regional audience. Under the stated policy, language or locale constraints should be explicit and justified or offered as an opt-in choice.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: This markdown file presents all user-facing instructions and descriptions exclusively in Chinese, with no indication that the skill is intentionally region-specific or that users may choose another language. Under the language/locale policy, forcing a specific language without opt-in can be a natural-language policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: This file contains user-facing natural-language content exclusively in Chinese, and there is no indication that users can opt into this language or that the content is limited to a Chinese-speaking audience. Under the policy, forcing a specific language without user choice can be a natural-language policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: This markdown file presents all user-facing content in Chinese and does not indicate that the language is optional, selectable, or limited to a China-specific audience. That can violate language/locale policy because it forces a specific language without user opt-in.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: This markdown file presents all instructions and content in a single language, which can constitute a language/locale policy violation when no user opt-in or scope justification is provided. The text does not indicate that the skill is region-specific or that Chinese output is optional.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The entire skill file is written in Chinese and provides no indication that language selection is optional or that the locale is intentionally restricted to a Chinese-only audience. Under the policy, forcing a specific language without user opt-in or documented justification is a natural-language policy violation.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 98% confidence
Finding: This file is materially off-scope for a ceiling-decoration skill: it focuses on air-conditioner type selection, aesthetics, and purchasing preferences rather than吊顶装修 guidance. In a specialized skill, irrelevant reference material can degrade retrieval quality and cause the agent to answer with unrelated recommendations, reducing trust and potentially producing bad renovation advice in context-sensitive decisions.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The article recommends potentially unsafe home modifications—such as enclosing outdoor-unit spaces, adding platforms, placing storage, pets, or laundry around AC outdoor-unit areas, and partially altering access zones—without safety warnings or a requirement for licensed professional review. These changes can obstruct maintenance access, affect ventilation/heat dissipation, create fall or load hazards, and potentially violate building/property rules, making the guidance risky if users act on it directly.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: This markdown file presents all instructions and content in a single language, Chinese, with no indication that the user can choose another language or that the skill is intentionally limited to a Chinese-speaking audience. Under the policy rule for natural-language violations, forcing a specific language without user opt-in is in scope.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: This markdown file presents all user-facing natural-language content in a single language and does not indicate that the user can choose another language or that the skill is intentionally limited to a Chinese-only region or audience. Under the language/locale policy, forcing a specific language without opt-in is a natural-language policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: This markdown file is entirely written in Chinese and provides no indication that the skill is region-specific or that users may choose another language. Under the policy, forcing a specific language without opt-in is a natural-language locale violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The entire file is written in Chinese and provides no indication that language selection is optional or that the content is intentionally limited to a Chinese-speaking or region-specific audience. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 84% confidence
Finding: The entire skill content is presented only in Chinese, with no indication that users may choose another language or that the skill is intentionally limited to a Chinese-speaking or region-specific context. Under the policy rules, language-specific behavior without opt-in or documented justification is a natural-language policy concern.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: This markdown file is entirely written in Chinese and provides no indication that language selection is optional or that the content is intentionally scoped to a Chinese-language or region-specific audience. Under the stated policy, forcing a specific language without user opt-in can be a natural-language policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The entire file is written in Chinese with no indication that language selection is optional or tailored to a region-specific audience. Under the stated policy, forcing a specific language without user choice is a natural-language policy violation.
Content: No source excerpt is available for this finding.