Blog

Insights

Curated posts from HappyMonkey — product builds, agent systems, and practical AI delivery. Daily grants/models/tooling roundups stay on the legacy WordPress site for now.

2026-08-29

Recent Work — A Roundup of HappyMonkey Projects

A look at the projects we've been building — from AI infrastructure and developer tools to interactive experiences and games.
2026-07-19

The night our MCP servers started working as a system

Last night I saw something I had been hoping for, but had not yet seen happen cleanly in a real project. Hermes needed to pass review findings to another coding agent, start a substantial piece of work, avoid overloading the local server, and keep enough
2026-07-16

GOSafe: location-based disaster risk intelligence for real places and dates

GOSafe (live at gosafe.happymonkey.ai) is our prototype for location-based disaster risk intelligence. It combines current events, official forecasts, and carefully bounded AI interpretation to answer a practical question: what should I watch around this
2026-07-16

OpenUKPublicDataMCP: one MCP endpoint for useful UK public data

Agents are only as good as the facts they can reach. OpenUKPublicDataMCP is a no-key-first Model Context Protocol server that gives Claude, Hermes, Cursor, and other MCP clients one reliable place to ask common UK public-data questions — with sources att
2026-07-16

miso-tts-docker: local Miso TTS 8B on Windows, Docker Desktop, and RTX 50-series

High-quality local speech is finally useful for agents and demos — if you can get the GPU stack to boot. miso-tts-docker is our unofficial Docker Compose pack for running Miso TTS 8B on NVIDIA GPUs, with a special focus on Windows, Docker Desktop, and RT
2026-07-16

Resource Sentinel MCP: stop local AI agents from melting the host

When you run many coding agents, tests, and models on one machine, the failure mode is boring and expensive: OOM kills, thrashing disks, and mysterious hangs. Resource Sentinel MCP is our answer — telemetry plus lease-based admission control that agents

2026-07-16

PromptOS: an AI-first Linux installer that never builds the same machine twice

Most installers ship a fixed menu of packages. PromptOS is different: a minimal live bootstrap that asks high-level questions, talks to your LLM provider, and generates a hardware-aware installation blueprint on the fly. The problem with static install m
2026-07-08

How Tokenization Powers (and Limits) Modern LLMs

When you type a prompt into ChatGPT, Claude, or Gemini, you’re not sending words. Your text is instantly broken into tokens — the fundamental units the model actually processes. Tokenization turns human language into the numerical units LLMs actual
2026-07-06

Building repo.spot: How We Used Qwen Cloud and Alibaba Function Compute to Create an Autonomous DevSecOps Agent

This post documents the journey of building repo.spot for the Global AI Hackathon Series with Qwen Cloud — an autonomous DevSecOps review agent powered by Qwen on Alibaba Cloud. The Problem Security vulnerabilities in open-source repositories are found,

2026-07-02

12 Simple Python AI Starter Projects for Beginners

Starting your AI coding journey does not mean jumping straight into deep learning. Here is a curated list of approachable Python projects—from classic ML exercises to lightweight apps you can ship with Streamlit—ordered from easiest to slightly more invo
2026-06-10

A2M update: local AI security audits, web submissions, and repository verification

I shipped some important new features for A2M today. The latest roadmap update is all about bridging the gap between local AI development, security, and marketplace visibility. 1. Local AI-powered security scanning The biggest change is a first-pass DevS
2026-06-10

Vibes TUI v0.1.3: a smarter AI agent harness with triage steering, stability fixes, and multi-provider support

Vibes TUI just got a meaningful upgrade. The latest release focuses on making the agent harness more resilient, more configurable, and easier to use in real work. What changed The headline feature is TriageAgent — a background observer that watches for t
2026-06-02

The Agentic Fork in the Road: Claude Code vs. Hermes Agent (And Why the Hype Has It Wrong)

If you spend ten minutes scrolling tech forums or AI newsletters right now, the consensus on terminal-bound AI coding assistants seems simple. You’ve got Anthropic’s shiny new Claude Code CLI on one side for engineering, and open-source autonomous framew
2026-05-25

Vibe Station: Visualising AI Workflows in Real Time

The latest Vibe Station updates are all about one thing: making AI work feel visible. When you’re building with AI, it’s easy to end up with a pile of prompts, logs, outputs, and disconnected tools. The result is a workflow that works, but doesn’t always
2026-05-24

We Built an AI Phone Assistant That Answers Calls, Captures Leads, and Follows Up Automatically

AI can already write code, summarize research, and automate inboxes — but one of the most practical business wins is much simpler: answering the phone well. We built an AI assistant that answers calls, speaks naturally, captures caller details in real ti
2026-05-24

ACE EU AI: Local-First Compliance for the EU AI Act

AI teams are building fast, but the EU AI Act deadline is looming. The problem? You can’t use cloud-reliant compliance tools to audit your system if your security policies bar you from leaking proprietary source code or model configurations outside your

2026-03-25

The Developer’s Dilemma 2025: Why Google AI Pro is the Full-Stack Winner

The Developer’s Dilemma 2025: Why Google AI Pro is the Full-Stack Winner As a developer in late 2025, the market is no longer just about “which AI is smartest.” It’s about which ecosystem actually solves the end-to-end development
2026-03-09

AI-Powered Personal Assistant MCP Server

AI-Powered Personal Assistant MCP Server Deploy a digital personal assistant that connects your calendar, Google Sheets, Gmail, and more—all orchestrated for ultimate productivity. This workflow aggregates meeting schedules, reminders, document tasks, an