A software factory that shows its working: a context layer for collaboration with frontier models, a loop engine powering autonomous workflows, and a structured development layer that turns long product roadmaps into executable slices.
Every new AI session used to start from zero: no memory of me, my projects, or how I work. The first ten minutes were always re-explaining yesterday.
Fulltrace is the answer: a personal operating system for the agentic era, one persistent memory that follows me across every tool, every session, and every project.
The foundation is the Context Portfolio: ten plain-text files describing who I am, what I am working on, and how I like things done. Every assistant reads the same files through a local MCP server; change one once and they all pick it up. Versioned in git, synced at the end of every session.
Live and in daily use. My assistants share one memory, the engine runs several kinds of job, and real changes have already shipped through it: to this site, and to my other projects' code once I have approved the exact change.
Fulltrace turns a product roadmap into shipped, verified code through a governed production line. Work enters as a slice, passes through multi-agent build, deterministic verification and adversarial review, and leaves with its full trace: who proposed it, what authority approved it, how it was checked, and what it cost to produce. Most AI coding setups generate code; Fulltrace manufactures trusted change, and the trace is the name.
A dated shipped register of over 120 slices, every one run kickoff to closeout under the same protocol. The build log narrates it; the roadmap is the live board.
Every change carries provenance and verification, not just a diff: run records, an append-only apply ledger, and gates enforced in code. The machinery lives on how it works.
Agents act under explicit, hash-bound grants with hard spend and write limits, so capability scales without scaling risk. The rules live on the workflow page.
Every assistant talks to one gateway: it serves the shared memory and runs the actual work. The portfolio files are the source of truth, git backs up everything, and the map shows the whole picture.
Fulltrace reviews this website the same way it reviews my other projects: it checks what the site claims against what the code shows, and ranks the fixes worth making first. Proven facts stay separate from guesses. The newest project through the loop is my Salesforce CI/CD work, with several review reports already produced. The mechanics live on how it works.
The full tour: where the shared memory lives, how the assistants plug into it, and how work runs, gets checked, and only then gets applied.
A control panel inside my code editor. I pick a job, see what it will cost before it runs, approve the spend, and watch it work in real time.
Three-layer config, auto-sync hooks, sub-agent delegation, and 17 shared slash commands.
Thirteen MCP tools; OpenRouter role profiles route planning, tooling, and agentic work pay-per-token.
Streamable HTTP to the shared server; command logging via hooks; shared jn-* skills deployed.
Started in April 2026 as a folder of notes about how I work. Piece by piece it grew: connections to each assistant, cost tracking, a dashboard, and finally an engine that checks and ships its own work. Every step is on the record.
read the build log →A personal system, built for one person, doing real work every day. Not a product, not a demo: infrastructure I rely on.