Back to skill

Security audit

casual-history-article

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only writing skill for Chinese casual history articles, with no hidden code, data access, or persistence beyond normal installation.

Install this if you want an agent to produce Chinese casual history or culture articles in a WeChat public-account style. In multilingual or general-writing workflows, configure the agent to use it only when that style is requested so it does not unnecessarily reshape neutral writing tasks.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill description mandates Chinese公众号-style output without indicating that this is conditional on user preference. If auto-selected, it can override the user's requested language or format, causing instruction conflict and unexpected output behavior in multi-skill agents.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The README describes activation in broad terms such as history, religion, geography,人物 and pop culture, with examples that can overlap many ordinary writing requests. In an agentic system, unclear trigger boundaries can cause the skill to activate when the user did not explicitly want this specialized behavior, leading to unintended steering of tone, structure, sourcing rules, or handling of sensitive topics.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The activation description is quite broad: it covers history/culture popularization, rewriting openings/endings, and matching copy for images, which can overlap with many generic writing requests. That can cause the skill to trigger when the user did not explicitly ask for this specialized behavior, leading to unintended instruction hijacking of tone, format, and workflow.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill description mandates Chinese output ('写成中文微信公众号文章') without indicating that this is conditional on user preference. If auto-activated, it may override the user's requested language or platform style, causing unauthorized instruction precedence and potentially degrading safety or usability in multilingual contexts.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

These changelog lines describe prescriptive writing behavior in Chinese, including formatting and stylistic rules, which may imply the skill enforces a specific language or locale. Because the file does not indicate that users can choose another language or opt in to this locale-specific behavior, it may conflict with language/locale choice policy.

Content

No source excerpt is available for this finding.