ClawHub
Back to skill

Security audit

Agentic Workflow Automation

Security checks across malware telemetry and agentic risk

Overview

This is a narrow workflow-blueprint generator; the main caution is that its dry-run option still writes an output file.

Reasonable to install for generating workflow blueprint files. Choose output paths carefully because files may be overwritten, and do not rely on --dry-run to avoid writes. Avoid putting secrets in workflow inputs because supplied content can be copied into generated artifacts.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill instructs the agent to run a local script and read a reference file, which implies file_read and likely code-execution/file-write capabilities, but the skill declares no permissions or capability boundaries. This creates a mismatch between what the skill appears to do and what users or enforcement layers are told it can do, increasing the chance of unintended file access or artifact creation without explicit approval.

Intent-Code Divergence

Medium
Confidence
97% confidence
Finding
The script advertises --dry-run as avoiding side effects, but it still writes the output file via render(), which violates caller expectations. In agentic or automation contexts, operators may rely on dry-run for safe preview behavior; unexpected file creation or overwrite can alter state, clobber artifacts, or trigger downstream automation that watches the output path.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.