Mac Mini AI — Mac Mini Local LLM, Image Gen, STT on Apple Silicon
v1.0.1Mac Mini AI — run LLMs, image generation, speech-to-text, and embeddings on your Mac Mini. M4 (16-32GB) and M4 Pro (24-64GB) configurations make the Mac Mini...
⭐ 2· 38·1 current·1 all-time
byTwin Geeks@twinsgeeks
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
The skill is an instruction-only guide for running a local fleet (herd/herd-node) and references installing a PyPI package (ollama-herd), running local HTTP endpoints, and using local model tooling — which matches the stated purpose. However, metadata marks python3/pip as optional while SKILL.md explicitly instructs 'pip install ollama-herd', so pip should be required; this mismatch is a minor coherence issue.
Instruction Scope
Instructions stick to setting up a local service, using local HTTP endpoints (localhost:11435), and interacting with models. They do not instruct reading unrelated system files or exporting data. Small scope ambiguities: the fleet 'automatic discovery' mechanism is not described (it may perform LAN discovery/multicast or open ports), and the 'uv tool install mflux' command is not explained (the 'uv' tool is undefined here). Also several examples call python3 utilities even though python3/pip were listed optional.
Install Mechanism
This is an instruction-only skill (no install spec), so it doesn't install code itself. It instructs the user to pip install a package from PyPI, which is a normal way to install CLI tools but does execute third-party code on install — users should inspect the package/repo before running. No arbitrary download URLs or archive extraction are embedded in the skill.
Credentials
The skill does not request environment variables or secret credentials and only references local config paths (~/.fleet-manager/*) which are relevant to the fleet. That proportionality is appropriate for the described functionality.
Persistence & Privilege
The skill does not request always:true, does not request system-wide changes in its instructions, and is user-invocable only. The skill starts local services (herd/herd-node) which is expected for this purpose but carries the usual runtime privilege of any local server process.
Assessment
This skill is largely coherent with its purpose, but check a few things before installing: 1) The SKILL.md tells you to run 'pip install ollama-herd' — inspect the PyPI package and its GitHub repo (https://github.com/geeks-accelerator/ollama-herd) to ensure the code matches expectations. 2) Confirm how the fleet 'discovery' works (it may open LAN ports or use multicast); if you need to limit network exposure, run in an isolated network or firewall the service. 3) The metadata lists python3/pip as optional even though examples use them — ensure you have a safe Python environment (virtualenv) before installing. 4) The 'uv tool install mflux' command is not explained here — verify what 'uv' is and where that tool comes from. 5) Running the herd will start local servers on port 11435 — review the service config and keep model downloads/installs manual as recommended. If these checks look good, the skill appears to do what it claims; otherwise treat it cautiously and inspect the code before installing.Like a lobster shell, security has layers — review code before you run it.
affordable-aivk9729wcvg0z4f6qdpvp45ta34n83zt20apple-siliconvk9729wcvg0z4f6qdpvp45ta34n83zt20embeddingvk9729wcvg0z4f6qdpvp45ta34n83zt20fleetvk9729wcvg0z4f6qdpvp45ta34n83zt20homelabvk9729wcvg0z4f6qdpvp45ta34n83zt20latestvk9729wcvg0z4f6qdpvp45ta34n83zt20llmvk9729wcvg0z4f6qdpvp45ta34n83zt20local-aivk9729wcvg0z4f6qdpvp45ta34n83zt20m4vk9729wcvg0z4f6qdpvp45ta34n83zt20m4-provk9729wcvg0z4f6qdpvp45ta34n83zt20mac-clustervk9729wcvg0z4f6qdpvp45ta34n83zt20mac-minivk9729wcvg0z4f6qdpvp45ta34n83zt20ollamavk9729wcvg0z4f6qdpvp45ta34n83zt20phi4vk9729wcvg0z4f6qdpvp45ta34n83zt20self-hostedvk9729wcvg0z4f6qdpvp45ta34n83zt20
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
computer Clawdis
OSmacOS
Any bincurl, wget