Distributed shared goal decomposition A goal goes in — “plan a two-week trip to Japan” — and a tree of executable prompts comes out, produced by four kinds of agent that never share a process, a database, or a coordinator. No peer ever holds the whole plan. Plan state is reconstructed locally by each peer. Intro I’ve been messing around with AI agents a lot recently, and two things I’ve built have been sitting in the back of my mind together for a while....
Intro I’ve been messing around with multi-agent AI systems recently. I had a crazy idea - what if I could get one AI agent to write code, have another score it, and a third refine it based on that score? All automatically. All in a loop. That’s what I’m going to walk through here. The things I wanted to explore were: Getting an AI agent to generate code from a prompt Getting a second AI agent to score that code and provide structured feedback Using that feedback to automatically refine the code in a loop Running the final accepted code as an actual subprocess TLDR - if you just want the code it’s here: https://github....
Intro I have been playing around with skills for a while now. I’ve also been using feature flags in my code for some time as well. Feature flags are a really neat way of separating deployment and release in any codebase. I decided to explore how I could use feature flags in a Claude skill. This would essentially enable me to change and control the skill using nothing other than feature flags....
Intro I’ve been messing around with AI and “agents” recently. I thought I’d share the architecture that I’ve come up with. I was working with some orchestration systenms and thought to myself “I want something that is autonomous, but also able to be completely distributed”. I have used the fantastic library libp2p. This gives me the ability to have peer discovery, a pub / sub message bus and more. I thought to myself “what if I combine my favourite network stack with the experiments I’m running with AI”...
Intro I have been fooling around a lot with ai recently, and one of the things I’ve found myself doing is integrating with MCP servers and then using the projects feature in claude to generate custom reports. What is this? I have set up a remote MCP server that pulls data from the Star Wars API (swapi). This is a remote MCP server, so I am hosting it on one of my domains....
Intro I have been fooling around a lot with ai recently, and I thought I would write something about what I’ve been doing. There are a few things that I’ve been doing, and they’re all fascinating. Writing average looking code using minimum viable prompts. Getting this average looking code refactored to be a robust service. Generating API documentation - and having Claude fix this code. This is part two of a small series that I have created to walk through the process I went through to get decent code....
Intro I have been fooling around a lot with ai recently, and I thought I would write something about what I’ve been doing. There are a few things that I’ve been doing, and they’re all fascinating. Writing average looking code using minimum viable prompts. Getting this average looking code refactored to be a robust service. Generating API documentation - and having Claude fix this code. This is part one of a small series that I have created to walk through the process I went through to get decent code....
Intro There has been a lot of hype on socials recently about MCP or Model Context Protocol. Essentially it’s a protocol that allows LLM AI to be extended. Most of the extensions relate to providing context using your own data. This is what I am going to walk through here. I will walk through: Creating an MCP What works and what works better My discoveries so far TLDR If you just want the code - it’s here:...