ClawHub
Back to skill

Security audit

PlantUML

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed local PlantUML diagram-rendering helper, with expected local command use and no evidence of credential access, persistence, exfiltration, or destructive behavior.

Install only if you are comfortable letting your agent render PlantUML locally with Java/Graphviz. Review .puml files from untrusted sources before rendering, avoid URL-based !include examples in sensitive or offline workspaces unless mirrored locally, and expect to adjust the example skill paths if your OpenClaw home directory differs.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill explicitly instructs the agent to execute local shell commands (`java -jar`, `python3`, `ls`) but does not declare corresponding permissions or constraints. That creates a capability/permission mismatch: users or the platform may treat the skill as documentation-only while it actually drives code execution against local tooling and potentially untrusted `.puml` input.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.