Tp2
High
- Category
- MCP Tool Poisoning
- Confidence
- 85% confidence
- Finding
Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
- Content
Security audit
SSQ Predictor — 双色球分析预测助手
Security checks for vulnerabilities and agentic risk
Overview
This is a single Markdown-only lottery analysis guide with no tools, code, network access, data collection, or persistence.
Install only if you want an entertainment-oriented Chinese 双色球 analysis prompt. Do not treat its generated numbers, historical-pattern claims, or update claims as reliable predictions or buying advice, and verify official rules and odds independently before spending money.
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 (4)
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 87% confidence
- Finding
The skill name and the entire operating description are written as a Chinese-language assistant, and the examples and workflow assume Chinese interaction. I did not find any statement that the user may choose another language or that Chinese is required for a region-specific compliance reason, which can violate the language/locale policy for natural-language behavior.
- Content
Intent-Code Divergence
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
The security notice explicitly says the skill 'does not contain executable code' and 'does not perform any actual lottery data collection or prediction computation.' However, later sections direct the agent to analyze metrics, filter numbers, validate combinations across multiple dimensions, and output 3-5 recommended tickets, which is active computational behavior rather than purely static reference material.
- Content
Intent-Code Divergence
Low
- Category
- Not specified by scanner
- Confidence
- 78% confidence
- Finding
The notice states there is no actual lottery data collection, yet the document later presents '2026年6月双色球数据更新' and says statistical model parameters were refreshed from official-source data. That creates an intent/documentation contradiction: either the skill relies on maintained draw data, or it should not present itself as having updated data-driven baselines.
- Content
Static analysis
No suspicious patterns detected.