agent-communicator

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:43 PM UTC
• Version:
  1.0.0

Overview

This is a markdown-only skill for formatting inter-agent messages and handoffs, with no code execution or hidden data access.

Use this as a formatting aid for agent-to-agent coordination. Review generated messages before sharing them, especially if they include private project context, credentials, customer data, or internal decisions; the skill itself does not run code or access those items.

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: 90% confidence
• Finding: The skill description uses broad activation language such as 'communication patterns' and 'quality assurance,' which can match many ordinary collaboration requests and cause the skill to trigger outside its intended scope. Over-broad invocation can lead to inappropriate routing, unintended disclosure of task context between agents, and increased prompt-surface for downstream prompt injection or authority confusion.

VirusTotal

64/64 vendors flagged this skill as clean.