Back to skill

Security audit

图书智能解读系统

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent book-analysis skill that creates local Markdown and HTML reports, with minor usability caveats around broad triggers, Chinese defaults, and optional CDN use.

Install this if you want Chinese-oriented book analysis reports. Expect it to create two local files per report, and clarify the output folder, overwrite behavior, preferred language, and whether online Mermaid rendering is allowed before using it on important material.

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)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
78% confidence
Finding

The README consistently describes invocation and outputs in Chinese, including trigger phrases and report naming, without stating whether users may choose another language. This can amount to a language-policy issue if the skill defaults users into a specific language without opt-in or documented justification.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The documented trigger phrases are very broad natural-language requests such as '分析/拆解某本书' and '这本书讲了什么', which can overlap with ordinary conversation and cause the skill to activate when the user did not explicitly intend to invoke it. In an agent environment, unintended activation can lead to unnecessary processing of user-supplied text and unanticipated generation of report artifacts.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill description uses broad trigger language such as requests to 'analyze/deconstruct a book' and generate summaries or reports, which can match a wide range of ordinary reading-related prompts without clear activation boundaries. This increases the chance of accidental invocation, causing the agent to apply this skill in contexts where a simpler response or a different skill would be more appropriate, potentially leading to scope overreach and unnecessary file-generation behavior.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill metadata and description are entirely Chinese-oriented and do not specify any mechanism for respecting the user's language preference. In a multi-lingual agent environment, this can cause the skill to respond in the wrong language or force a language choice inconsistent with the user's request, reducing reliability and potentially exposing confusing or unusable output.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger scenarios enumerate positive examples but do not define exclusions, precedence rules, or minimum evidence that the input is actually a book-analysis task. Because the skill can activate on just a title, excerpts, notes, or recommendation-style prompts, ambiguous requests may incorrectly route here, increasing the risk of misexecution and overbroad handling of user content.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The HTML report规范 explicitly requires lang="zh-CN", which imposes a fixed language/locale choice. Under the policy, locale constraints should either be optional, user-selected, or clearly justified as region-specific; this file provides no opt-in or justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The README states that the skill outputs '<书名>-解读报告.md' and '<书名>-解读报告.html' but does not say where files are written or warn the user that local artifacts will be created. In agentic systems, silent file creation can surprise users, clutter workspaces, or overwrite expected files if naming collisions are not handled safely.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

该技能清单前文将产出描述为 Markdown + HTML 双格式解读报告,且自查要求 HTML 版“无外部依赖、可直接双击打开”。但这里又写明在某些情况下可嵌入 mermaid.js(CDN:cdn.jsdelivr.net),这引入了可选的外部网络依赖,与离线分享定位存在语义冲突。

Content

No source excerpt is available for this finding.