Back to skill

Security audit

orangeink

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent Chinese WeChat Markdown publishing skill with disclosed, mostly user-directed file output and no hidden credential handling or persistence.

Install only if you need Chinese WeChat public-account formatting. Use normal rendering with explicit input/output files, avoid --no-check for publishable output, do not enable OIMD_REMOTE_TEMPLATE unless you trust the upstream template, and prefer a local pinned index.html for parity checks. Do not render or open copy pages from untrusted Markdown containing raw HTML or scripts.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • 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
Findings (13)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 97)May include surrounding context.

md
- **结构合规**:本 skill 即 Claude Agent Skills 通用规范布局(`SKILL.md` + `scripts/` + `references/` + `tests/`),frontmatter 仅用通用字段(name / description / license / metadata)—

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
88% confidence
Finding

The CLI accepts a --no-check flag that disables the static compliance/security-style validation before output generation and exit-code signaling. In an automated publishing pipeline, this can let non-compliant or unsafe HTML pass through silently, weakening a defense-in-depth control and making downstream misuse or policy bypass easier.

Content

Scanner excerpt · scripts/render.mjs (reported line 988)May include surrounding context.

js
'--meta': v => { args.meta = v; },
  };
  const BOOL = { // 布尔开关:不消费后续参数
    '--quiet': () => { args.quiet = true; }, '--no-check': () => { args.noCheck = true; },
  };
  for (let i = 2; i < argv.length; i++) {
    const a = argv[i];

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.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The description states the skill is for the Chinese WeChat public-account scenario and that its output is Simplified Chinese. This imposes a fixed language policy rather than offering a user choice or opt-in, which matches the locale-policy violation criteria.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The locale metadata specifies zh-CN and says the product definition is Chinese output, which is a natural-language policy constraint on language/locale. Because no user opt-in or alternative is offered in the metadata, it fits the policy-violation category.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The title and description specify producing content in a particular Chinese writing style for WeChat public-account Markdown, and the rest of the file assumes Chinese output conventions. This is a natural-language locale constraint, but the file does not present it as an opt-in choice or explicitly justify the restriction as a region-specific tool policy.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This code includes natural-language comments and user-facing operational text in Chinese, and later emits Chinese error/help messages to stderr/stdout. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation unless the tool is clearly documented as locale-specific, which is not evident in this file.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The script supports fetching index.html from raw.githubusercontent.com and then executes that HTML's embedded scripts inside jsdom with runScripts: 'dangerously'. Even though this path is gated behind OIMD_REMOTE_TEMPLATE=1, it creates an explicit remote code retrieval-and-execution path, which is risky because any compromise of the upstream repository, transport assumptions, or trusted content would result in code execution in the local analysis environment.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The thrown error text is presented only in Chinese and is intended for end users running the script. Because the file does not offer a language selection mechanism or justify a Chinese-only audience, this violates the language/locale policy criterion.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

This runtime failure message is displayed directly to the user and is only available in Chinese. The policy requires either user choice of language or clear documentation justifying the locale restriction, neither of which appears in this file.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

This code generates a copy page with lang="zh-CN" and all visible UI text in Chinese, which imposes a specific language/locale in the user-facing output. The file does not present any opt-in, fallback, or configurable locale selection, so it violates the language-choice policy for natural-language behavior.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The comments state that remote fetching is disabled by default and should not be used in audit scenarios, but the feature remains implemented and can be enabled via environment variable. This is dangerous because operators may rely on the comment as a safety guarantee while the code still contains a latent remote execution path that can be activated by configuration, increasing the chance of accidental unsafe use.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The visible user-facing content is entirely in Chinese, including the heading metadata and introductory text, with no indication that language is optional or region-specific. Under the policy rule, a skill that imposes a specific language without user opt-in can be a natural-language policy violation.

Content

No source excerpt is available for this finding.