Back to skill

Security audit

彩票测算器

Security checks for vulnerabilities and agentic risk

Overview

This skill openly runs a local lottery-divination script to generate entertainment lottery numbers, with no evidence of hidden access, persistence, network use, or data exfiltration.

Install only if you want a Chinese-language entertainment lottery-number generator. Treat the output as random entertainment, not financial advice, and be aware that the skill may activate on broad Chinese number-picking phrases.

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 (4)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The trigger definition is overly broad and explicitly includes casual phrases like '帮我算个号', making accidental or contextually unrelated activation likely. This can cause the agent to invoke the skill without clear user intent, overriding normal conversational flow and increasing the risk of inappropriate responses in unrelated contexts.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

Mandating Chinese-only output without considering the user's language can cause unsafe or unusable interactions, especially if the user cannot understand disclaimers, gambling-risk warnings, or clarifications. In this skill, the disclaimer is a required safety control, so forcing a single language weakens the effectiveness of that protection for non-Chinese-speaking users.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This Python file includes its primary description, usage guidance, and user-facing output text in Chinese only. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation, and there is no indication that users may choose another language or that the skill is intentionally region-restricted.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
98% confidence
Finding

The comments repeatedly assert different, incompatible line encodings for the same trigrams, including remarks like 'No!', 'wait', and contradictory mappings for 巽/兑/震, before settling on a final table. This is more than incomplete documentation: the inline documentation actively disagrees with itself and with intermediate code states, making developer intent unclear for a divination engine whose output depends on exact trigram definitions.

Content

No source excerpt is available for this finding.