T08 · Insecure Dependencies
Warning
- Location
SKILL.md:247- Finding
Unpinned Third-Party Dependencies Installed from an Unspecified Package Source
- Content
View full analysis
- Remediation
View remediation
Security audit
ZJTJ-SAR指标计算器
Security checks for vulnerabilities and agentic risk
Overview
This is a disclosed stock-analysis helper, but users should treat its buy signals as rough educational output rather than financial advice.
Install only in an isolated Python environment, prefer pinned dependencies, and do not treat the generated recommendations as trading instructions. The strategy is simplified and partly inconsistent with its own scoring description, so verify any result independently before making financial decisions.
Vulnerability Patterns
- Insecure DependenciesIntroduces malicious components through unsafe dependency sources
- 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
Findings (1)
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)
Description-Behavior Mismatch
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The manifest and documentation describe a four-dimensional stock-picking strategy with dimensions including 市场情绪 and a total-score threshold of >=60. However, the implemented analyze_stock logic only checks ZJTJ approximation, SAR trend, and volume ratio, then recommends based on having at least two conditions, with no market-sentiment input or score calculation.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 92% confidence
- Finding
The skill presents '建议买入' style outputs without a clear investment-risk disclaimer or statement that it is not financial advice. In the finance context, this increases the chance that users over-trust simplified heuristics and act on recommendations without understanding limitations, suitability, or risk.
- Content
Intent-Code Divergence
Medium
- Category
- Not specified by scanner
- Confidence
- 91% confidence
- Finding
The documented rule says the SAR dimension is satisfied by '上涨趋势 + 价格在SAR上方'. In the actual selection logic, analyze_stock records price_above_sar but does not use it when appending the SAR condition; it adds 'SAR上涨趋势' solely when trend == 1.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The trigger phrases are broad enough to activate on ordinary finance-related conversation, which can cause unintended invocation of the skill. In this context, that matters because the skill produces stock screening and buy recommendations, potentially influencing users without explicit intent to use this tool.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 86% confidence
- Finding
文件整体描述、输出标签、触发短语和示例均默认且仅使用中文，没有说明该技能仅面向中文用户，也未提供语言切换或用户选择。若组织要求避免无用户同意的语言强制，这构成潜在的语言/locale 策略问题。
- Content
Static analysis
No suspicious patterns detected.