Back to skill

Security audit

攒够了去旅行

Security checks for vulnerabilities and agentic risk

Overview

This skill is a lightweight Chinese travel-points reward assistant with no executable code, persistence hooks, credential access, or hidden destructive behavior.

Installers should expect a Chinese-language, motivational travel-points helper. Be aware that its broad matching may award points for vague statements, and clicking the included preview links takes the user to an external web page not inspected as executable package code.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (8)

Vague Triggers

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The "general_high_energy" behavior uses non-specific phrases like "不错" and "挺好", which are common conversational expressions, and the note states that vague expressions are classified here by default. This creates an overly broad trigger condition that can capture ordinary speech without a clear boundary for activation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The configuration sets "default_language" to "zh-CN" and does not indicate any user choice or opt-in mechanism. Per the policy, forcing a specific language or locale without offering user selection is a natural-language policy concern unless clearly justified as region-specific.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This JSON manifest defines activation-style keywords for behaviors, so SQP-1 applies. The keyword list for "important_task" includes very broad everyday terms like "完成", "搞定", "做完", and "提交", which overlap with common speech and lack scope constraints or negative examples, making unintended matches likely.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This markdown file specifies the product principles exclusively in Chinese, and nowhere indicates that users may choose another language or that the skill is intentionally limited to a Chinese-speaking audience. Under the language/locale policy, forcing a specific language without user opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

L005-L013 明确说明产品的核心承诺包括“打开就能拿到一组真实可做的奖励建议”和将小奖励逐步攒成旅行;而技能清单描述只允许在记录行为、换算旅行积分、查看目标差距或积分记录时触发,不包含奖励建议生成。该文档体现的实际产品能力明显超出了清单声明的触发范围与用途。

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

L104-L109 写明用户可“按类型抽”“按状态匹配”“添加到奖励池”;L125-L138 进一步定义了添加奖励时系统要理解用户输入、判断类别并放入奖励池。这些是独立的奖励管理能力,不是清单所述的积分累计/进度查看的显然实现细节。

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

The markdown explicitly states '默认中文', which sets a language default for responses. Under the policy, language constraints should offer user choice or be clearly justified; this file does not mention opt-in, fallback behavior, or a region-specific justification.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
75% confidence
Finding

L017-L023 断言产品不是“自律打卡系统”,但文档后文围绕“刚完成一件事”的状态匹配、添加奖励池以及累积解锁旅行展开,实际仍依赖对用户行为完成情况的记录与奖励机制。虽然不是直接代码行为,但该文档意图与技能围绕记录高能量行为的实际定位存在张力,容易误导实现和使用边界。

Content

No source excerpt is available for this finding.