Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md node scripts/run.mjs --sample # 先看样例
Security audit
采购返利与阶梯核算核对（免费版）
Security checks for vulnerabilities and agentic risk
Overview
This is a local purchase-rebate checking skill with no network, credential, persistence, or data-modifying behavior, though users should note its free tier does not perform some advertised full-tier validations.
Install only if you are comfortable running a local Node-based checker for Chinese purchase-rebate tables. Do not rely on it for bracket-rate validation, tax/accounting judgment, or final financial decisions unless you independently verify the omitted checks and contract terms.
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 (7)
Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md node scripts/run.mjs --sample # 先看样例
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 90% confidence
- Finding
The manifest description says '触发词包括 采购返利核对、返利没收足、阶梯返利、年度返利', but several of these phrases are short natural-language business terms that could appear in ordinary discussion rather than as explicit skill-invocation commands. The file does not provide negative examples, context limits, or a precise activation rule to distinguish when these phrases should trigger the skill versus when they are merely mentioned.
- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
The file’s natural-language description, user-facing advice, and result messages are all written only in Chinese, with no indication that users may choose another language or locale. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation.
- Content
Intent-Code Divergence
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
The file states that withheld checks are only descriptive text and not implemented, but later code still consumes bracket-related input associated with those withheld checks. This documentation/behavior mismatch is dangerous because it can cause operators, wrappers, or automated systems to trust an analysis scope that is different from what the code actually signals, leading to incorrect business decisions.
- Content
Description-Behavior Mismatch
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The free-version engine advertises that rate-bracket matching is withheld, yet it still accepts a brackets input and reports brackets_used in the output. Even though the brackets are not actually applied to validate rates, exposing and echoing this configuration can mislead downstream users or agents into believing bracket-based checks were performed, creating a silent integrity failure in a financial verification workflow.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The JSON contains only Chinese-language text with no indication that the skill offers a language choice or that it is intentionally limited to a Chinese locale. Under the policy rule for natural-language violations, forcing a specific language without user opt-in can be a locale-policy issue.
- Content