Local Llm Ops

This is a batteries-included toolkit for running Ollama on macOS with Apple Silicon. It bundles setup scripts, a CLI chat launcher with task modes (coding, creative, analytical), benchmarking harness, and diagnostics into one repo. The workflow is straightforward: run setup_chatbot.sh, pull a model like mistral, then launch with ./chatllm or use aliases for quick access. Benchmarks are scripted and repeatable, which is handy if you're comparing models or tracking performance across updates. The diagnostic script is a nice touch when things break. If you're running local LLMs regularly and want the operational scaffolding already built, this saves you from stitching together your own scripts.

Install to Claude Code

npx -y skills add bobmatnyc/claude-mpm-skills --skill local-llm-ops --agent claude-code