Ai Agent Orchestration Advisor

• Outcome: Pass
• Latest audit: May 28, 2026 at 5:49 AM UTC
• Version:
  3.0.0

Overview

This is a text-only advisor for choosing multi-agent AI frameworks, with no hidden execution or data access found.

Install this if you want help comparing multi-agent frameworks and designing agent workflows. Be aware it may trigger on broad agent-framework discussions, and verify its framework, cost, and benchmark claims against current official sources before making technical or spending decisions.

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: 93% confidence
• Finding: The trigger word list is very broad and contains many generic framework names and phrases without clear invocation boundaries. This can cause the skill to activate in unrelated conversations, leading to unintended influence over agent decisions, prompt-space collisions with other skills, and increased exposure to prompt injection through accidental routing.

VirusTotal

62/62 vendors flagged this skill as clean.