Back to skill

Security audit

Diagram Drawing

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed diagram generator that creates local SVG/PNG files, with minor caution around broad activation phrases and output paths.

Install this if you want an agent to generate diagram files locally. Be explicit about when you want a diagram and where outputs should be written, especially in shared or important directories.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (5)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill declares very broad trigger phrases such as 'visualize', 'create chart', and generic Chinese phrases like '帮我画'/'生成图', plus it triggers on 'any system/flow description.' This can cause unintended activation in normal conversation, leading the agent to switch into file-generating behavior unexpectedly and potentially execute follow-on tool actions the user did not clearly request.

Content

No source excerpt is available for this finding.

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/icons.md (reported line 22)May include surrounding context.

Agent / Orchestrator (hexagon)

xml
<!-- r=circumradius. For r=36: points at 36,0  18,31.2  -18,31.2  -36,0  -18,-31.2  18,-31.2 -->
<polygon points="cx,cy-r  cx+r*0.866,cy-r*0.5  cx+r*0.866,cy+r*0.5  cx,cy+r  cx-r*0.866,cy+r*0.5  cx-r*0.866,cy-r*0.5"
         fill="FILL" stroke="STROKE" stroke-width="1.5"/>
<text x="cx" y="cy+5" text-anchor="middle" fill="TEXT" font-size="12" font-weight="600">Agent</text>

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill instructs writing SVG/PNG files to disk and even provides a Python snippet that writes to an arbitrary path, but it does not require user consent, path confirmation, or safety checks. In an agent environment with file system tools, this can lead to unexpected persistence, overwriting files, or writing outputs to sensitive or unintended locations.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/svg2png.py (reported line 31)May include surrounding context.

python
cmd.insert(2, str(width))
    
    try:
        result = subprocess.run(cmd, capture_output=True, timeout=30)
        return result.returncode == 0
    except (subprocess.TimeoutExpired, Exception):
        return False

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
70% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · scripts/svg2png.py (reported line 116)May include surrounding context.

python
Install one of:
  1. rsvg-convert (recommended):
     macOS:    brew install librsvg
     Ubuntu:   sudo apt install librsvg2-bin
     
  2. cairosvg (Python):
     pip install cairosvg

Static analysis

No suspicious patterns detected.