Department Manager

• Outcome: Pass
• Latest audit: May 28, 2026 at 11:44 AM UTC
• Version:
  1.0.0

Overview

This skill appears to be a local department and task organizer for agents, with no evidence of network access, exfiltration, hidden execution, or destructive behavior.

Installers should treat the skill as a local task-state manager: avoid putting secrets, credentials, or highly sensitive business details into task text or completion output unless local storage at ~/.openclaw/department-manager/departments.json is acceptable.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Vague Triggers

Medium

• Confidence: 80% confidence
• Finding: The manifest description is very broad and could cause the skill to be selected for many generic planning, delegation, or project-management requests. Overbroad activation increases the chance that a user or orchestrator invokes a skill with code-executing and state-modifying behavior in situations where a simpler, lower-privilege skill would have been safer.

Vague Triggers

Medium

• Confidence: 82% confidence
• Finding: The introductory text frames the skill as a general-purpose way to organize AI workers 'like a CEO' for autonomous businesses or complex projects, without meaningful constraints. In context, this makes the skill more dangerous because it is user-invocable and paired with command examples that create, assign, complete, and remove persisted task state, so loose triggering can expand operational scope unnecessarily.

Missing User Warnings

Low

• Confidence: 89% confidence
• Finding: The script persists department names, task text, and completion output to a local JSON file under the user's home directory without notifying the user or offering a non-persistent mode. In this skill's context, task and output fields may contain sensitive business plans, credentials, internal prompts, or agent-generated data, so silent persistence increases the risk of unintended local disclosure to other local users, backups, or later compromise of the host.

VirusTotal

64/64 vendors flagged this skill as clean.