Back to skill

Security audit

长期目标管理

Security checks for vulnerabilities and agentic risk

Overview

This skill transparently maintains a local goal.md file for long-running work and includes a local validator, with no evidence of hidden execution, exfiltration, or unrelated privilege use.

Install this only if you want the agent to keep a persistent goal.md in your workspace for long-running tasks. Review the chosen goal.md path before first use, especially in repositories where adding tracking files matters, and expect the provided validator/template to require Chinese section names and fields.

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 (9)

Tp4

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding

声明描述的是一个长期目标管理技能,核心能力应是维护和延续 goal.md 作为长期工作的持久化状态载体。但给出的代码并不管理、写入、合并或延续任何长期目标状态,而是一个针对 goal.md 模板结构的验证脚本。它的主要功能是格式/完整性校验,包括章节是否齐全、字段是否唯一、时间格式是否合法、表格子目标 ID 是否重复、占位符是否未替换,以及文件中的“权威目标”路径是否指向自身。虽然这与“唯一 goal.md 作为单一事实来源”的概念有关,算是支持长期目标工作流的配套工具,但其主要目的明显是验证而非维护。因此代码行为只覆盖了声明中的很小一部分支持性需求,且主用途与声明的“管理长期目标”存在实质偏差。

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding

The skill clearly instructs the agent to read, create, and repeatedly update a persistent goal.md, but it declares no explicit tool scope or permission boundary. That makes file-system side effects implicit rather than transparent, increasing the chance of unexpected workspace modification or broader-than-necessary file access when the skill is invoked.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger conditions are broad enough to match common conversational phrases like 'continue', 'long-term', 'iterative', or 'from handoff materials', which can cause accidental invocation. In this skill, accidental invocation is more dangerous because it leads to persistent state restoration and file writes to a canonical goal.md, potentially altering project state without clear user intent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill repeatedly instructs the agent to create, update, and validate goal.md, yet it does not clearly warn users up front that invoking the skill will modify workspace files. Hidden or insufficiently disclosed file writes are risky because users may trigger the skill expecting planning assistance, while the agent persistently changes repository state and may create new project artifacts.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The default prompt broadly instructs the agent to use the long-goal skill whenever a task appears long-term, but it does not define firm activation boundaries, exclusions, or confirmation requirements. This can cause the skill to be invoked in contexts where persistent state, goal tracking, or carryover from prior sessions is unnecessary or sensitive, increasing the risk of unintended data retention, context contamination, or inappropriate continuation from stale goal material.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This code requires specific Chinese section headings such as '## 总目标' and Chinese field labels to pass validation, effectively forcing one language/locale. The file does not indicate user opt-in or document a justified region-specific constraint, which matches the language/locale policy violation criteria.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The required labels include Chinese-only metadata fields like '状态:', '最近维护:', '权威目标:', and others, so documents in other languages would be rejected. Because the script provides no alternative locale support or explicit user choice, this is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

该文件整体以中文规定技能行为和输出要求,但没有说明这是可选语言,也没有提供根据用户偏好切换语言的机制。若组织要求避免无用户选择地强制特定语言,这种单一语言约束构成自然语言策略风险。

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

This markdown template is written entirely in Chinese and provides no indication that another language can be used or selected. Under the policy rule for natural-language violations, a fixed language requirement without user opt-in can be a locale/language policy concern.

Content

No source excerpt is available for this finding.