Fulltrace is how I put AI to work. A software factory builds my roadmap day and night, task runs check and fix my projects, and a chat in my editor answers quick questions. Plain code checks the work, and every step leaves a record I can replay.
Demo available on desktop or tablet only.
From the most hands-off to the most hands-on. Each one has its own page if you want the detail.
The factory builds Fulltrace from its own roadmap, one small planned piece of work at a time. I call each piece a slice. Every slice starts with a written plan that names its checks, and ends with a written closeout and a commit.
Low and medium risk slices run with nobody in the session, one after another, day and night. A slice that needs my decision is set aside so the queue keeps moving. If a slice changes the runner's own files, the chain stops until I check.
Riskier slices change how the system writes, approves or is set up. They run with me in the session, and so does design work. They need a written design before any code.
I choose the order of the queue, and the Claude model and how hard it thinks for each level of risk.
Audits and fixes across all my projects, run from Studio in my editor. Before a run starts, Studio estimates its cost from past runs, and I watch every step live. Plain code checks each result, and a change to my files waits for my approval of that exact change. Runs can also be scheduled overnight, inside limits I set.
I choose the job, the projects, and a model for each stage: planning, doing, checking and fixing.
A chat inside my editor that can use any model on OpenRouter, or one running on my own computer at no cost. It has its own API key, and every reply shows which model answered and what it cost, under a spending limit. It cannot see my files or notes, so it cannot change anything.
I choose the model, and how hard it thinks when the model supports it.
Each way has its own models, its own checks and its own record. The factory and task runs share one local server, which also briefs my coding assistants. Chat stands apart from both.
Each column, top to bottom: how the work starts, the models it uses, how it is checked, and what it leaves behind.
Every new AI session used to start from zero, with no memory of me, my projects or how I work. The first ten minutes went on re-explaining yesterday.
Now Claude Code, Codex and Cursor all read the same ten plain-text files about who I am, what I am working on and how I like things done, through one small server on my machine. I change a file once and every assistant picks it up. The files are kept in git and synced at the end of every session.
Where it all started: completing the AI Daily Brief AgentOS program.
Live and in daily use. The factory ships slices most days, task runs have made real fixes to my projects' code once I approved the exact change, and a website loop keeps this site current.