Sam

gollama. Go manage your Ollama models

ingest. Parse files (e.g. code repos) and websites to clipboard or a file for ingestions by AI / LLMs

delete-unknown-root-ca. Delete Known Government-Linked Certificate Authorities in OS X

llamalink. Link you Ollama models to LM-Studio

mcp-devtools. A modular MCP server that provides commonly used developer tools for AI coding agents

agentic-coding. Agentic Coding Rules, Templates etc...

mcp-package-version. An MCP server that provides LLMs with the latest stable package versions when coding

mcp-package-docs. An MCP server that provides LLMs with efficient access to package documentation across multiple programming languages

mcp-llm. An MCP server that provides LLMs access to other LLMs

superprompter. Supercharge your AI/LLM prompts

llm-templates. My LLM Templates (Ollama Modelfiles & Tabby Templates + Presets)

homeassistant-localai. LocalAI integration component for Home Assistant

kernel-ci. Continuous integration for the Linux Kernel - Built within Docker

the_best_of. At the end of every year I do a bit of summary of the best applications, hardware & websites I've enjoyed & depended on throughout the year (and often for some time before).

n8n-nodes-searxng. n8n community node of searxng

moa. Mixture-of-Ollamas

vlm-ui. Web Interface for Vision Language Models Including InternVLM2

zsh-bootstrap. bootstrap my zsh shell

scripts. misc scripts

bybit-mcp. Ollama MCP server for bybit

NVApi. A lightweight API that returns Nvidia GPU utilisation information.

mcp-snyk. MCP Server for Snyk Security Scanning

firecrawler. A lightweight frontend for self-hosted Firecrawl instances

mcp-github-issue. An MCP server that provides LLMs with the ability to use GitHub issues as tasks

urldiff. Computes and displays the visual differences between two URLs

quantest. LLM (v)RAM estimator for GGUF models from huggingface and ollama across various quantisation and context sizes

gomcp. Go

ollama-artefacts. Build HTML artefacts with Ollama

storage. Open Source, Code Controlled, Enterprise Ready SSD Storage

packer-qemu-debian-jessie. Packer qemu template for Debian 8 (Jessie)

smart_diskinfo. Script to gather SMART info on all block devices

spitter. Copies Ollama models between hosts

parameter-playground. TypeScript

mcp-data-extractor. A model context protocol server to migrate data out of code (ts/js) into config (json)

homekit_device. Home Assistant Device Aggregation for Homekit

quant. My visualisations and understanding of LLM quantisations

LDAP-Self-Service. Web Based LDAP Self Service Portal

Ender-3-V2-Neo. My Documentation, Configuration and Scripts for the Ender 3 V2 Neo 3d Printer

esphome-esp-s3-box-3-volume. add volume controls to the esp-s3-box-3 in esphome

tabload. TabLoad is a GUI application for interacting with TabbyAPI the Exllamav2 Server

ender-configs. Ender Printer Configuration, Scripts and Notes

bedrock-openai-api. OpenAI Compatible API for Amazon Bedrock

github-app-installation-token. Generates Github tokens using a Github App install

sqs-message-action. Github Action to publish an AWS SQS message

sammcj. JavaScript

esphome-esp-s3-box-3.

ai-artifacts. Hackable open-source version of Anthropic's Claude Artifacts

m2e. Sick of seeing Zs where there should be Ss? This app converts text from 'Murican to International English

open-webui-pipelines. My Open WebUI Pipelines (Functions, Filters, Tools)

llamanator. AI/LLM middleware between Home Assistant and Ollama

getcitoken. Retrieves Gitlab-CI Token for Automated Runner Authentication

airqualitylogger. Logs data from HT2000 air quality monitors to CSV

CO2-Logger. Monitor CO2 concentration using MH-Z19 (or the older Z14) sensor connected with UART.

VibeVoice4macOS. VibeVoice on Apple Silicon (macOS) — no CUDA

local-deep-research. Local Deep Research is an AI-powered assistant that transforms complex questions into comprehensive, cited reports by conducting iterative analysis using any LLM across diverse knowledge sources including academic databases, scientific repositories, web content, and private document collections.

actions-dashboard. Github Actions Dashboard

renovate-config. My renovate config

MoA-Ollama-Chat. This is the Mixture-of-Agents (MoA) concept, adapted from the original work by TogetherAI. My version is tailored for local model usage and features a user-friendly Gradio interface.