Qwen Qwen3 5

• Outcome: Pass
• Latest audit: May 28, 2026 at 1:59 PM UTC
• Version:
  1.0.1

Overview

This is a documentation-only skill for running a local Ollama Herd Qwen setup, with the main caution being its reliance on an external PyPI package and local service endpoints.

Before installing, review the `ollama-herd` PyPI package and repository because `pip install` executes third-party code. Run the router only on networks you intend to trust, and treat the dashboard/status endpoints and ~/.fleet-manager logs as local operational data that may reveal usage details.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• 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)

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The skill is marketed and named as a Qwen 3.5 model-serving skill, but the documentation also exposes unrelated capabilities such as other LLMs, image generation, embeddings, monitoring endpoints, and dashboard access. This expands the operational scope beyond user expectation and can lead agents or users to invoke features with different risk profiles, especially local service and observability endpoints that may disclose data or enable unintended actions.

VirusTotal

VirusTotal findings are pending for this skill version.