ClawHub
Back to skill

Security audit

draw architecture diagram

Security checks for vulnerabilities and agentic risk

Overview

This skill is a design template for dark architecture diagrams and does not request unusual access or perform hidden actions.

Install this if you want a strong preset style for dark architecture diagrams. Review generated HTML before sharing if language/locale matters, and remove the Google Fonts link if you need a fully offline or no-network artifact.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (8)

Hidden Instructions

High
Category
Prompt Injection
Confidence
70% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 47)May include surrounding context.

```

Hidden Instructions

High
Category
Prompt Injection
Confidence
70% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · references/component-patterns.md (reported line 31)May include surrounding context.

The primary structural element. Each architectural layer gets a boundary rect with its accent color.

svg
<!-- Layer boundary: two overlapping rects for stroke + fill -->
<rect x="40" y="{y}" width="1120" height="{h}" rx="12"
      fill="none" stroke="{layer-color}" stroke-width="1.5" stroke-opacity="0.3"/>
<rect x="40" y="{y}" width="1120" height="{h}" rx="12"

Hidden Instructions

High
Category
Prompt Injection
Confidence
70% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · references/component-patterns.md (reported line 220)May include surrounding context.

L-Shape Route (avoid crossing)

svg
<!-- Horizontal first, then vertical, then horizontal -->
<line x1="{start-x}" y1="{start-y}" x2="{corner1-x}" y2="{start-y}"
      stroke="{color}" stroke-width="1.2" stroke-opacity="0.5" stroke-dasharray="5,3"/>
<line x1="{corner1-x}" y1="{start-y}" x2="{corner1-x}" y2="{corner2-y}"

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The example HTML structure hard-codes the document language to Simplified Chinese, which can impose a specific locale on all outputs generated with this skill. The surrounding instructions do not offer a language choice or explain that the skill is intended only for a China-specific context.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The file includes fixed Chinese labels like "LLM 推理域" and "（可能多轮循环）", and elsewhere specifies an English-only subtitle. This imposes presentation language expectations in natural language without indicating that users may choose their preferred locale or that the document is intentionally region-specific.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The legend hard-codes the label "执行步骤" for the step indicator. Because this is natural-language UI text in a general component pattern file, it should not force a specific language unless the file clearly states a region- or locale-specific purpose.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The placeholder explicitly says "Technical subtitle in English," which is a direct language requirement. For a general-purpose component pattern document, forcing English presentation text can violate language-choice policy when no opt-in or scope justification is provided.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The font stack explicitly prioritizes 'PingFang SC', a Simplified Chinese font, as the fallback after JetBrains Mono. This creates a locale-specific rendering preference in a natural-language design specification without offering alternatives or documenting user choice.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.