I built a coding workflow called Plumbline (https://github.com/BytesFromToby/plumbline) — five AI-agent "skills" that take a feature from idea to verified code. The names lean on a home-building metaphor on purpose: an architect defines what to build, a foreman plans how, a builder writes the code, an inspector signs off. A plumb line is the weighted string a builder hangs to find true vertical — the one reference everything else is checked against. In my workflow, the spec is that line. One way to wire this is a chain: architect calls foreman, foreman calls builder. I deliberately didn't. The agents never call each other. They coordinate entirely through files and folders — and that one choice turned out to be the most important in the whole thing. Chaining them causes two problems: tight coupling, where a failure halfway means restarting the whole pipeline; and — worse with LLMs — a telephone game, where each agent paraphrases the last one's output until the builder is working from a summary of a summary of what I actually asked for. # The fix: the filesystem is the interface Every stage reads and writes plain files in a known layout: Planning/specs/[feature]_spec.md ← architect writes this; it's the source of truth Planning/blueprints/[feature]_BP.md ← foreman writes the build plan output/inspect/ ← inspector drops evidence + a PASS/FAIL stamp A stage's input is whichever files already exist; its output is whatever it writes. No call graph. The builder runs after the blueprint exists — sequenced by data dependency, not because the foreman told it to. A single CLAUDE.md in each project declares the shared contract (test command, where specs live, the rules for a change) — the schema that lets independently-run agents agree on where everything is without ever talking. # Why it's better than I expected Every stage runs standalone — the pipeline is a suggestion, not a cage. Intent stops drifting: every stage reads the original spec, not a retelling of it, so the builder checks against what I actually asked for and stops if the plan contradicts it. Independence becomes structural: my inspector only reads the finished files and never saw the build, so its neutrality is enforced by isolation, not good intentions. And everything is inspectable — when something breaks I read the files, not a black-box conversation.

Hey Everyone, Just wanted to share a small win for me today. I managed to secure my first sit down with a possible client today. I got talking to our contract manager at work today just general chit chat talking about his role and what it entailed. We got onto how late he has to work and how it was hard on the family so so forth. I said to him what if i could give you that time back through implementing software that did a lot of the heavy lifting when it comes to office admin. he laughed and asked how was that I explained lv learned how to use AI to build systems we went back and forth a bit about AI talking about where it might lead and such and I'm not going to lie basically regurgitated Jakes classes stating things like how its now to have a hand carved rifle will set you back a small fortune but way back when you would of been looked down on, I spoke about the part about human desire and how that will always be there. He was completely taken back to him I was now not just a vet and a carpenter I was the future and Iv been asked to come to the office tomorrow to sit down and talk. I know its only a sit down and nothing confirmed but when i first approached them i was basically laughed out the office and given work via what's app but that small amount of knowledge might be the reason my life changes for the better. So thank you @Jake Van Clief .for giving this course away and helping complete beginners like me a fighter chance of a better future.