Ae1
High
- Category
- analysis-evasion
- Content
node scripts/run.mjs --sample # 先看样例（一张干净台账，0 条结论）
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Security audit
在建工程转固与利息资本化核对（免费版）
Security checks for vulnerabilities and agentic risk
Overview
This is a local Chinese-language accounting ledger checker with no evidence of network use, persistence, credential access, or hidden behavior.
Install only if you are comfortable running a local Node-based checker on Chinese-language construction-in-progress ledger text. It does not appear to send data out or persist anything, but it is not a substitute for accounting or audit judgment and it documents several checks as out of scope.
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 (5)
Ae1
High
- Category
- analysis-evasion
- Content
node scripts/run.mjs --sample # 先看样例（一张干净台账，0 条结论）
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Natural-Language Policy Violations
Medium
- Confidence
- 91% confidence
- Finding
- The user-facing name, summary, description, and trigger terms are presented almost entirely in Chinese, and the input instructions later assume Chinese column names. Although there is a brief English display name/description, the skill does not state that Chinese is required or offer an explicit language/locale choice, which can violate language/locale policy expectations.
Natural-Language Policy Violations
Medium
- Confidence
- 91% confidence
- Finding
- This .js file contains natural-language instructions, sample data, labels, advice, and result messages entirely in Chinese, and the runtime outputs in lines later in the file also assume Chinese input and Chinese responses. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale constraint is explicitly justified or user choice is offered.
Intent-Code Divergence
Low
- Confidence
- 85% confidence
- Finding
- The inline documentation says non-JSON input is treated as plain-text material such as a full contract document, which broadens the apparent input scope beyond the manifest's stated purpose of checking CIP ledger items, capitalization, and transfer-to-fixed-asset details. The code indeed accepts arbitrary text files as input, so the comment suggests a wider use case than the skill's stated intent.
Static analysis
No suspicious patterns detected.