Draft Agent Loop

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:39 PM UTC
• Version:
  1.6.0

Overview

This skill is a disclosed Draft-based approval workflow, but users should understand that task journals are stored and potentially published through Draft.

Install this only if you want task plans and progress recorded in Draft. Avoid putting secrets, private personal data, proprietary code dumps, or sensitive command output in the Task Journal, and review the Draft CLI dependency and invite-code settings before using it for confidential work.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (1)

Vague Triggers

Medium

• Confidence: 76% confidence
• Finding: The trigger condition 'high-stakes or complex enough to warrant human gates' is subjective and overly broad, which can cause the skill to activate without an explicit user request for this workflow. In an agent system, broad activation criteria can override user intent, route tasks into remote page publication flows unexpectedly, and increase exposure of task metadata to external systems.

VirusTotal

65/65 vendors flagged this skill as clean.