T08 · Insecure Dependencies
Warning
- Location
SKILL.md:22- Finding
Unpinned Third-Party Dependencies Installed from a Mutable Package Index
- Content
View full analysis
- Remediation
View remediation
Security audit
A Stock Screener
Security checks for vulnerabilities and agentic risk
Overview
This A-share stock-screening skill is purpose-aligned and not harmful by itself, but users should treat its output as rough research because several checks are simplified.
Install only in a virtual environment, review the dependency sources, and treat generated stock lists as preliminary research rather than investment advice. Confirm any result against official disclosures and current market data 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)
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 90% confidence
- Finding
The description explicitly defines the skill as an 'A 股五步量化选股技能' and all user-facing examples and instructions are in Chinese, but there is no note that the language is optional or limited to a China-specific audience. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy concern.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The trigger phrases are very broad, generic natural-language requests such as '帮我筛选 A 股' and similar variants. In an agent environment, these can cause unintended skill invocation during ordinary conversation, leading the agent to switch into stock-screening behavior without clear user intent or consent.
- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 91% confidence
- Finding
This markdown file is entirely written in Chinese and presents the guidance as the default output with no indication that users may choose another language or locale. Under the policy rule, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is explicitly documented and justified.
- Content
Intent-Code Divergence
Medium
- Category
- Not specified by scanner
- Confidence
- 97% confidence
- Finding
The script promises a concrete five-step quantitative stock-screening process, but several steps are implemented as placeholders, arbitrary head() truncation, or simplified examples rather than the documented checks. In an investment decision-support skill, this can mislead users into believing the output is evidence-based and risk-screened when it is not, potentially causing financial harm through reliance on false analytical rigor.
- Content
Intent-Code Divergence
Low
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
The CLI exposes an --industry option that is never applied, creating a mismatch between user intent and actual screening behavior. In a financial screening context, users may act on results believing they were filtered for a target sector when they were not, undermining trust and potentially leading to incorrect portfolio decisions.
- Content
Static analysis
No suspicious patterns detected.