Back to skill

Security audit

专注时段与习惯排程助手

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to be a disclosed focus and habit scheduling helper, with optional AI-HIVE use and no hidden persistence, credential collection, or destructive behavior found.

Install only if you want a Chinese-language scheduling helper and are comfortable with optional AI-HIVE use. Do not paste passwords, tokens, cookies, verification codes, or broad private datasets into it; review any upload scope, price, budget, and external action before authorizing.

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 Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

声明描述的是一个时间管理/习惯打卡/专注排程类助手,核心能力应是接收用户任务与时间约束并产出排程和清单。实际代码却是通用的离线验证器:读取结果JSON和输出契约,检查记录状态、动作类型、授权引用、证据、domain字段状态,以及相对路径工件的哈希和大小限制,并输出valid/errors。代码注释还明确说明“never authenticates, calls APIs, or executes actions”,进一步表明它不是排程助手。两者在主要目的、输入输出和核心能力上均明显不一致,因此应判定为描述与行为不匹配。

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest description says the skill should be used when the user needs '专注助手、习惯打卡、时间管理Agent、拖延待办助手', which are broad everyday productivity phrases rather than narrowly scoped invocation triggers. It does not provide explicit trigger constraints or negative examples in the invocation description, increasing the risk of unintended activation.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The manifest uses broad task-oriented activation language such as completing focus/habit planning from user materials, but it does not define clear boundaries for when the skill should or should not be invoked. This can cause overbroad routing or unintended activation, increasing the chance that the agent is used outside its intended scope and handles user requests or dependencies without sufficient qualification checks.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The metadata is written to enforce Chinese-language interaction without documenting that this is a locale-specific skill or offering language fallback. This can misroute users, reduce informed consent about language handling, and create operational errors if users provide materials in another language or expect a different locale.

Content

No source excerpt is available for this finding.