Mlx Apple Silicon Mlx

Security checks across static analysis, malware telemetry, and agentic risk

Overview

This is a coherent local-MLX setup guide, but it asks users to install external Python tools and run a local multi-device router service.

This skill appears benign and purpose-aligned for running a local MLX AI fleet on Apple Silicon. Before installing, verify the external Python packages, run the router only on trusted devices/networks, and be mindful that prompts, audio, and generated content may pass through the local fleet service.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Installing these packages will run code from external package repositories on the user’s Mac.

Why it was flagged

The skill instructs installation of external, unpinned packages. This is purpose-aligned for a local AI stack, but the package code is not included in the reviewed artifacts.

Skill content
pip install ollama-herd ... uv tool install mflux ... uv tool install diffusionkit
Recommendation

Install only if you trust the package sources; consider checking the PyPI/GitHub project, versions, and dependencies before running the setup.

What this means

A local AI service may keep running and coordinate with other devices until the user stops it.

Why it was flagged

The skill tells users to run long-lived router and node processes. This is disclosed and central to the fleet-routing purpose, but it affects the local runtime environment.

Skill content
herd # start the router (port 11435)
herd-node # run on each device — finds the router automatically
Recommendation

Run the router and nodes only when needed, understand how to stop them, and use them on trusted devices/networks.

What this means

Data sent to the local API, such as prompts or audio files, may be processed by the local router and connected nodes.

Why it was flagged

The skill describes a multi-device local fleet router. That is expected for this tool, but prompts, audio, or generated workloads may be routed through local services/devices.

Skill content
One fleet router coordinates all four across Mac Studio, Mac Mini, MacBook Pro.
Recommendation

Use the fleet only with trusted local machines and avoid sending sensitive files unless you are comfortable with how the local service handles them.