Back to skill

Security audit

Novel Publish Ready

Security checks for vulnerabilities and agentic risk

Overview

This is a Chinese novel quality-checking skill that runs a local text scanner and gives editing guidance; its behavior is mostly disclosed and proportionate.

Install this only if you want Chinese manuscript review and are comfortable letting the agent read the chapters you provide and create a local freq_result.txt scan report. Consider keeping sensitive drafts in a limited folder and invoking the skill explicitly for the files you want checked.

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
93% confidence
Finding

声明描述的是一个范围很广的小说质量审查技能,涵盖结构、逻辑、风格与角色设定等高级分析能力,并声称有固定执行流程。实际代码只是一个静态文本规则扫描工具,核心能力限于词频统计、逐行关键词匹配、句式标记和替换建议。它确实部分符合“违禁词”“词频扫描”的子功能,但没有实现声明中的大部分主要能力,且也没有展示完整报告生成、自检复核或章节级综合质量分析流程。因此描述明显夸大了代码实际行为,属于能力与用途上的实质性不匹配。

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
83% confidence
Finding

The skill requires use of a local script and iterative scanning/report generation behavior, implying file read/write capabilities, but it declares no explicit tool scope or permission boundaries. In an agent system, undeclared capabilities increase the risk of overbroad filesystem access, accidental data exposure, or execution against unintended files because reviewers and runtime policy cannot clearly constrain what the skill may touch.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger keywords are very broad and include common editorial terms like '自检', '大纲', and '人设', which can cause the skill to activate in unrelated contexts. Over-triggering is dangerous in agent environments because it may cause unsolicited file access, analysis of sensitive drafts, or execution of auxiliary tooling when the user did not intend to invoke this skill.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

The rule “英文残留 | 中文替代” imposes a Chinese-only language policy in natural-language instructions. The file does not offer user opt-in, alternative locale handling, or a documented region-specific justification for forcing this language constraint.

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 file contains user-facing natural language in the module docstring that assumes Chinese as the required interaction language. Under the policy, language constraints should not be forced unless the user is given a choice or the restriction is clearly documented and justified as region-specific.

Content

No source excerpt is available for this finding.