Security audit

瓦匠全流程助手

Security checks for vulnerabilities and agentic risk

Overview

This is a masonry helper skill with disclosed calculators and reference guidance, and I found no hidden persistence, credential access, network execution, or destructive behavior.

Install this if you want a Chinese-language masonry calculation and reference assistant. Verify prices, safety rules, waterproofing practices, and structural advice against local codes or a qualified professional, especially for leaks, falling tiles, load-bearing walls, or high-risk repair work.

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

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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

Findings (26)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 95% confidence
Finding: The code does implement part of the declared description: material calculation for bricks, mortar, tiles, grout, adhesive, and plaster, plus some mortar ratio data and tile layout-related quantity estimation. However, the description claims a much broader full-process assistant with 8 major modules, including quotation reference, construction guidance, quality diagnosis, acceptance inspection, safety guidance, and interactive HTML visualization reports. None of those capabilities appear in the supplied code. The actual code is narrowly focused on quantity calculations and JSON output via command line. Some declared scenarios such as waterproofing and floor leveling are also not implemented in this chunk. Therefore, the description materially overstates the behavior of the code provided.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The trigger set is broad and contains common construction terms, increasing the chance the skill activates for loosely related conversations. Unintended invocation can cause prompt-context hijacking of normal chats, leading to irrelevant guidance, misleading calculations, or HTML report generation when the user did not request this skill.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: This Python file contains natural-language descriptions, comments, help text, and output labels entirely in Chinese, including the main module docstring and CLI usage guidance. Because the skill does not offer any language or locale selection and does not explain that it is intentionally region-specific, it violates the language/locale policy criteria for natural-language content.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 88% confidence
Finding: The trigger '瓦匠' is short and generic within the domain, so the skill may activate on incidental mention rather than explicit intent to use the assistant. This can inject domain-specific instructions into unrelated conversations and reduce reliability of agent routing.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 88% confidence
Finding: The trigger '瓦工' is overly short and likely to appear in ordinary discussion about construction workers. That raises the probability of accidental activation and inappropriate response shaping.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The natural-language content, examples, outputs, and role instructions are entirely in Chinese, which effectively forces a language choice on users. There is no statement that the skill supports multiple languages, asks for user preference, or justifies a China-specific locale constraint.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 90% confidence
Finding: The trigger '砌墙' is a common phrase and may match broad home-improvement discussions without a request for this specific skill. Such accidental routing can result in misleading calculations or procedural advice being surfaced out of context.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 90% confidence
Finding: The trigger '贴砖' is broad and commonly used in renovation conversations. This increases the chance of unintended invocation, which can interrupt normal chat flow and provide construction guidance when not solicited.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 90% confidence
Finding: The trigger '瓷砖' is very generic and likely to occur in many unrelated shopping, design, or repair contexts. A generic noun as a trigger materially increases accidental activation risk.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 85% confidence
Finding: The trigger '空鼓' is short and may appear in general defect discussions outside the desired skill flow. That can route users into this skill without confirming they want masonry diagnosis assistance.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 87% confidence
Finding: The trigger '抹灰' is a common construction term and is too short to reliably signal deliberate skill invocation. The security concern is unintended activation rather than direct exploitation.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 91% confidence
Finding: The trigger '防水' is especially broad and can appear in many non-masonry contexts, including high-consequence home repair discussions. Misrouting could produce overconfident construction guidance or implied professional advice without enough context, which is more concerning here than for narrower terms.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 84% confidence
Finding: The trigger '勾缝' is short and may be mentioned casually in tile or finishing discussions. Its breadth makes accidental activation plausible, though impact is limited to routing/reliability issues.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 86% confidence
Finding: The trigger '砌筑' is broad domain terminology rather than a clear intent signal. This creates unnecessary activation surface for the skill.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 87% confidence
Finding: The trigger '掉砖' may activate during urgent defect/safety complaints where users are describing a hazard rather than invoking a skill. In that context, accidental activation is somewhat more dangerous because the assistant may provide simplified repair guidance instead of escalating to immediate safety advice.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 89% confidence
Finding: The trigger '渗水' is generic and can occur in serious leak or structural discussions outside the skill's intended scope. Unintended activation could cause premature troubleshooting advice where urgent mitigation or professional inspection is needed.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 86% confidence
Finding: The trigger '砌砖' is too short to unambiguously identify a request for this skill. As with other broad triggers, the main issue is accidental activation and degraded agent routing.
Content: No source excerpt is available for this finding.

Overly Broad Trigger

Low

Category: Trigger Abuse
Confidence: 86% confidence
Finding: The trigger '铺砖' is common and underspecified, making false-positive activation likely in ordinary renovation conversations. The resulting risk is mainly user confusion and context hijacking.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 89% confidence
Finding: This markdown file contains user-facing natural language exclusively in Chinese, and it does not state that the skill is region-specific or that users can choose another language. Under the language/locale policy, forcing a specific language without opt-in can be a policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 93% confidence
Finding: This markdown file contains only Chinese-language content and does not offer an alternative language, user opt-in, or a stated region-specific justification. Under the policy rule for natural-language violations, forcing a specific language without user choice can be a locale-policy issue.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 95% confidence
Finding: This markdown file is natural-language reference content, and all headings and instructions are presented only in Chinese. Under the policy rule, forcing a specific language without user opt-in can be a locale/language policy violation when no justification or alternative is provided.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 90% confidence
Finding: SQP-3 applies to all file types and covers language or locale policy violations. This file presents all guidance in a single language with no opt-in, alternative language option, or justification that the skill is intended only for Chinese-speaking users or a China-specific context.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 82% confidence
Finding: SQP-3 applies to all file types and includes language or locale policy violations. This markdown file presents all safety guidance solely in Chinese, with no indication that users may choose another language or that the restriction is justified by a region-specific requirement.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 96% confidence
Finding: This markdown file contains user-facing natural language exclusively in Chinese, and there is no indication that the skill offers alternative language support or that the Chinese-only constraint is intentional and documented. 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

Low

Category: Not specified by scanner
Confidence: 93% confidence
Finding: This file presents all instructional content in a single language and does not state that the skill is China-specific or otherwise limited to Chinese-speaking users. Under the policy, forcing a specific language without user opt-in can be a natural-language locale violation.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.