[{"data":1,"prerenderedAt":111},["ShallowReactive",2],{"navigation":3,"/blog/4-omega-settings-one-document":26,"/blog/4-omega-settings-one-document-surround":107},[4],{"title":5,"path":6,"stem":7,"children":8,"page":25},"Blog","/blog","blog",[9,13,17,21],{"title":10,"path":11,"stem":12},"How to run LLMs on your local machine?","/blog/1-introduction-to-ollama","blog/1-introduction-to-ollama",{"title":14,"path":15,"stem":16},"Hosting a website on a Raspberry Pi - Part 1: why bother?","/blog/2-pi-hosting-part-1-why","blog/2-pi-hosting-part-1-why",{"title":18,"path":19,"stem":20},"omega: an AI agent platform that runs on your machine","/blog/3-introducing-omega","blog/3-introducing-omega",{"title":22,"path":23,"stem":24},"omega today: settings as one document","/blog/4-omega-settings-one-document","blog/4-omega-settings-one-document",false,{"id":27,"title":22,"author":28,"body":30,"date":92,"description":93,"extension":94,"image":95,"meta":96,"minRead":103,"navigation":104,"path":23,"seo":105,"stem":24,"__hash__":106},"blog/blog/4-omega-settings-one-document.md",{"name":29},"Endo",{"type":31,"value":32,"toc":88},"minimark",[33,37,40,43,46,54,57,60,63,66,69,82,85],[34,35,36],"p",{},"omega shipped 0.17 this week. The changelog undersells it, so here is where the project actually is. Two things: the settings tab stopped being a settings tab, and the way this thing gets built has quietly become the most interesting part.",[34,38,39],{},"First the boring-sounding one.",[34,41,42],{},"Settings in agent tools are usually a pile of dialogs. omega had the same disease. Toggles scattered across panes, knobs that lived in three places, no honest answer to \"what did I change\".",[34,44,45],{},"0.17 rebuilds all of it as one document. A search bar that actually filters. Commonly used knobs first, everything else in one scrollable list, grouped by what they do. A dot marks what you changed from default. One click puts it back. Same shape VS Code settled on years ago. That shape won for a reason.",[34,47,48,49,53],{},"The agents tab got the same treatment. Every agent now has a real chat bound to its latest session, so you talk to the agent you configure, not to a roster row. And agent settings render as per-agent scopes: a ",[50,51,52],"span",{},"Main | agents"," strip, and each agent tab shows only the things an agent can override - provider, model, system prompt, max turns. Nothing else. The config layering law is the same one the binary boots on.",[34,55,56],{},"Now the part I actually want to talk about.",[34,58,59],{},"I do not write omega's code. An agent fleet does. One main agent plans the work with me. Worker agents implement it, one card at a time. A reviewer agent gates every card before it merges - no card lands unreviewed. And every night a test suite drives a real browser through the web UI like a user would: dock a pane, send a chat, break the socket, watch it reconnect.",[34,61,62],{},"The receipts: 1,808 Go tests, 1,759 web UI tests, both green on the release tip. The settings rebuild above went through that pipeline as six cards in one day.",[34,64,65],{},"Here is why you might not care. This pipeline took real law to build. Agent workers will happily merge their own work if a card body has a gap, so every rule they must follow lives in the repo, not in vibes. The docs drifted too - a recent audit found nine claims that were flat false against the actual code. All nine were written by agents. All nine got fixed by one card. Agent-built software still needs someone who owns the merge. That is me. Every commit to master, every tag, every release is my call, and it stays that way.",[34,67,68],{},"The models ride Ollama Cloud endpoints. The product is local-first - one binary, loopback by default, your hardware - but I am not going to imply the build farm runs on a potato. It runs on a Pi, a Windows box and a cloud bill.",[70,71,72,76,79],"ul",{},[73,74,75],"li",{},"The socket reconnect that never happened: close events were going unnoticed and a dead socket stayed dead. The browser journeys caught it, a measured fix landed, the journey now goes green.",[73,77,78],{},"The docs got a truth pass: every claim was checked against the tree, not against memory.",[73,80,81],{},"The composer lost its workdir selector - it lives in the status bar now, where it always belonged.",[34,83,84],{},"The next window is the context arc. We measured a real inefficiency in how the prompt gets assembled - the tool schemas were reshuffling every single turn, which quietly wrecks prompt caching - and the fix is already through review. Compaction and context selection follow.",[34,86,87],{},"One binary, settings that read like a document, and a fleet that ships when the tests say so. 0.17 is on GitHub now. The roadmap is public.",{"title":89,"searchDepth":90,"depth":90,"links":91},"",2,[],"2026-11-12T00:00:00.000Z","State of the project: v0.17 turns settings into one searchable document, the agents tab gets a real chat, and the fleet that builds omega is the part worth watching.","md","/blog/omega-0-17.png",{"tags":97,"ogImage":95},[98,99,100,101,102],"AI","Agents","Go","Self-hosted","Development",3,true,{"title":22,"description":93},"oJBRV2WRxxa1WJQ94Nn6MzdogOV1u8pCcOAs71t7veo",[108,110],{"title":18,"path":19,"stem":20,"description":109,"children":-1},"Chat, tools, sessions, cron, kanban - an agent workspace as one binary you own. No cloud, no accounts, no telemetry. Local-first by default.",null,1794487578742]