ClawHub
Back to skill

Security audit

Solo Workflow Engine

Security checks across malware telemetry and agentic risk

Overview

This skill is mostly a no-code automation guide, but it asks for broad local command and file permissions that are not well scoped for that purpose.

Review this skill before installing if your agent can execute local commands. It appears intended as automation-planning guidance, but you should require command previews and explicit approval for any shell execution, and avoid granting broad write or exec permissions unless you need them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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 (4)

Context-Inappropriate Capability

Medium
Confidence
91% confidence
Finding
The skill is presented as a workflow methodology/configuration guide, but its error-handling instructions tell the agent to run network diagnostics such as ping. That expands behavior from advisory guidance into host-level command execution without clear necessity, scope limits, or user confirmation, creating risk of unintended command execution and capability overreach.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill advertises exec capability in its tool set without a prominent warning that it may execute host commands or describing the trust boundary. In an agent environment, exposing command execution without explicit user-facing safeguards can lead users to invoke the skill without understanding that local system actions may occur.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The example tells the agent to run a network diagnostic command during troubleshooting but does not warn the user that a shell command will be executed on the host. Even a seemingly harmless command normalizes host command execution and can be repurposed or extended in unsafe ways in a skill that already exposes exec.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The usage example includes shell command execution as part of a normal success path without discussing system impact, permissions, or the need for user acknowledgment. This makes command execution appear routine for a mostly documentation-oriented skill, increasing the chance of unreviewed host actions.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.