I'd be curious to hear more about your work on the 'Sovereign AI Stack'. I'm also working on a project that prioritizes governance and verification and I'd love to compare notes.
Importing git history is ugly but do-able. I had to do that at a previous job (splitting a git repo in two pieces or importing commit history from SVN).
I can take a look and try to create a PR around this if there is interest.
My father still has one of these in orange and white. I remember when I was a little child and he would start it up, I could feel the concussion of the exhaust in my chest.
I've been neck deep in a personal project since January that heavily leverages LLMs for the coding.
Most of my time has been spent fitting abstractions together, trying to find meaningful relationships in a field that is still somewhat ill-defined. I suppose I could have thrown lots of cash at it and had it 'done' in a weekend, but I hate that idea.
As it stands, I know what works and what doesn't (to the degree I can, I'm still learning, and I'll acknowledge I'm not super knowledgeable in most things) but I'm trying to apply what I know to a domain I don't readily understand well.
I've found that value is largely derived from polish and vision.
It's easy to prompt some stuff into existence over a weekend. It is hard to polish it, fix bugs, have tidy UX, and so on. There's this meme going around (maybe from that Silicon Valley show?) where the grey-beard says he is valued for his taste and his conviction in that taste. This is -- fortunately or not -- reality.
Vision and taste won't get you the whole way, but they are a huge part of the equation. This is why Apple, for example, was so successful under Jobs: he had vision, and he had good taste.
I agree, and for those who would counter “just use AI to polish”, those who use AI to avoid doing the work of building something are likewise going to avoid doing the work to polish it, if they even possess the taste required to do so.
I've been working on a framework since the end of January or so. I'm on my 7th draft. As I've gone along, each draft gets markedly smaller. The overlaps between what I'm building and openclaw are significant, but I've realized the elements that make up the system are distinct, small, and modular (by design).
There are only a few primitives:
1. session history
1a. context map + rendered context map (think of a drive partitioning scheme, but for context -- you can specify what goes into each block of context and this gets built before being sent out for inference).
2. agent definition / runtime
3. workflow definition / runtime
4. workflow history
5. runtime history (for all the stuff session and workflow history fail to capture because they are at a lower level in the stack)
That's it. Everything else builds on top of these primitives, including
- memory (a new context block that you add to a context map)
- tool usage (which is a set of hooks on inference return and can optionally send the output straight back for inference -- this is a special case inside the inference loop and so just lives there)
- anything to do with agent operating environment (this is an extension of workflows)
- anything to do with governance/provenance/security (this is an extension of either workflows and/or agent operating environment... I haven't nailed this down yet).
I suppose I should say something about how agents and workflows work together. I've broken up 'what to do' and 'how to think' into the two primitives of 'workflow' and 'agent' respectively. An agent's context map will have a section for system prompt and cognitive prompt, and an agent can 'bind' to a workflow. When bound, the agent has an additional field in their context map that spells out the workflow state the agent is in, the available tools, and state exit criteria. Ideally an agent can bind/unbind from a workflow at will, which means long-running workflows are durable beyond just agent activity. There's some nuance here in how session history from a workflow is stored, and I haven't figured that out yet.
Generally, the idea of a workflow allows you to do things like scheduled tasks, user UI, connectors to a variety of comms interfaces, tasks requiring specific outputs, etc. The primitive lays the foundation for a huge chunk of functionality that openclaw and others expose.
It's been fun reasoning through this, and I'll admit that I've had an awful lot of FOMO in the mean time, as I watch so many other harnesses come online. The majority of them look polished, and are well marketed (as far as AI hype marketing goes). But I've managed to stay the course so far.
I hope you find your ideal fit. These tools have the potential to be very powerful if we can manage to build them well enough.
reply