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
发票领用存与开票数据核对（免费版）
Security checks for vulnerabilities and agentic risk
Overview
This skill is a local invoice ledger checker that reads a user-provided file and does not show hidden network, persistence, credential, or destructive behavior.
Install only if you are comfortable running a local Chinese-language Node checker on invoice usage and stock ledger text. It does not appear to send data out or persist anything, but users should understand that the free version reports some summary rates while not performing the paid abnormal-rate checks.
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
- 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
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)
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
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
This code's natural-language interface is entirely in Chinese, including the main file description and operational guidance, and later user-facing error/disclaimer strings also assume Chinese input/output. The file does not offer any language choice or document a justified region-specific constraint, which matches the locale-policy violation criteria.
- Content
Intent-Code Divergence
Low
- Category
- Not specified by scanner
- Confidence
- 88% confidence
- Finding
The top-of-file comment states that CHECKS_WITHHELD is only explanatory text and not implementation. Yet later code calculates void_rate, reverse_rate, and exposes the corresponding caps, which constitutes partial implementation of the withheld rate-based checks even if no explicit finding is emitted.
- Content
Description-Behavior Mismatch
Low
- Category
- Not specified by scanner
- Confidence
- 91% confidence
- Finding
The manifest and file header say the free edition only implements free checks, and CHECKS_WITHHELD are merely not-executed items. However, the result scope still computes and returns void_rate, reverse_rate, and their caps, which are the core inputs for the withheld abnormal-rate checks listed in CHECKS_WITHHELD.
- Content
Intent-Code Divergence
Low
- Category
- Not specified by scanner
- Confidence
- 92% confidence
- Finding
The comment says non-JSON input is treated as plain-text material such as a pasted '合同全文' (full contract), which contradicts the surrounding code and manifest that are narrowly scoped to 发票领用存台账核对. The implementation and CLI help consistently expect invoice usage/stock ledger text with headers, so this documentation example misstates the intended material type.
- Content