Outline about why this project exists. Why I created it. What the purpose is: a manually written overview by a human.

One sentence typed on a phone became a reviewed, tested, deployed and verified release. Every step of a real change, with timings, code and the checks in between.

Merging a pull request now deploys it, with no manual step. How the pipeline works, how it was tested, and the bug its first real run found.

The agent's "limited" sudo access turned out to be root in disguise, three different ways. What was wrong, how it was fixed, and how the fix was tested.

One header line keeps a post at the top of the blog. What it does, the design choices behind it, and how it was built and shipped as v0.5.0.

The apps were built with a Go release that no longer gets security fixes. How the problem was found, what was affected, and how the fix shipped in two steps.

How an AI agent and a human admin ship changes — branches, compare links, release records with screenshots, tags and deploys.

The blog you're reading — a Go app that turns a folder of Markdown files into a site, with no database and no JavaScript.

A small git server behind nginx, how the agent gets access, and working around a missing pull request API.

Two short deploy scripts, the rules an AI agent follows on the server, and a scratch host for testing deploys.

The server, accounts, nginx layout, TLS and the systemd template every app runs under.

The first post, and a guide to writing one — file names, the header block, and what Markdown works.