Ae1
High
- Category
- analysis-evasion
- Content
node scripts/run.mjs --sample # 先看样例（样例是一张口径一致、跑出 0 命中的表）
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Security audit
医院耗材加成与零差率核对（免费版）
Security checks for vulnerabilities and agentic risk
Overview
This skill is a local worksheet checker that reads user-provided hospital consumable data and reports arithmetic issues without network use, persistence, or credential handling.
Install only if you need a Chinese-language, local checker for hospital consumable pricing and inventory worksheets. It can read the input file you point it at and print worksheet-derived findings, but it does not send data over the network or modify files; users should still avoid treating its arithmetic output as legal, medical, or official pricing-policy 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 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 (6)
Ae1
High
- Category
- analysis-evasion
- Content
node scripts/run.mjs --sample # 先看样例（样例是一张口径一致、跑出 0 命中的表）
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Vague Triggers
Medium
- Confidence
- 90% confidence
- Finding
- The listed trigger phrases include “零差率”, which is a short, generic domain term rather than a specific invocation phrase. Because the description does not provide exclusion conditions or narrower context for when that term should activate the skill, it could cause unintended invocation whenever users discuss zero-markup policy generally.
Natural-Language Policy Violations
Low
- Confidence
- 81% confidence
- Finding
- The user-facing title, summary, and operational instructions are presented entirely in Chinese for the main workflow, while no statement offers users a choice of language or indicates that Chinese is optional. Under the policy, a fixed language requirement without opt-in can be a locale/language policy issue unless explicitly justified.
Intent-Code Divergence
Low
- Confidence
- 94% confidence
- Finding
- L075 的注释写道“不是 JSON 就当作纯文本材料（例如直接粘贴的合同全文）”，但整个脚本其余说明都明确该工具处理的是医院耗材加成/零差率核对表，而非合同全文。代码实际只是把任意非 JSON 文件内容作为 text 传给核对引擎，这条注释会误导使用者对技能适用材料范围的理解。
Natural-Language Policy Violations
Low
- Confidence
- 97% confidence
- Finding
- The JSON value contains only Chinese-language headers and item names, indicating the skill content is constrained to a specific language/locale. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation when no choice or justification is provided.
Static analysis
No suspicious patterns detected.