The compose files, the gateway's Caddyfiles, the egress proxy's configuration, the example
.env, the seccomp profile, the dev-server's example service definitions and the stack's two
documents moved to mcps-docker-compose. With the images published to Docker Hub, a server
runs them without the sources, and a deploy is a new image tag; keeping both here meant the
repository that changes with the code also carried what only changes when a host does.
publish-images.sh builds the four images for amd64 and arm64 and pushes them tagged with
this repository's commit hash and "latest". It refuses a dirty working tree, since the
tag names a commit. The images now published were built from 921d54c.
Each server keeps its README and tests, and coding-agent-mcp its own standalone compose
files. The three references that pointed at moved files now say where they went.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
They were three folders on one laptop: a coding agent with a git history of its
own, and two servers - a development server and a browser - with none at all.
What makes them a stack is what sits between them, and that lived nowhere: the
compose file, the gateway, the egress proxy and the networks that keep the worker
off the internet. So the whole thing is one repository, and the coding agent's
separate history is folded into it rather than kept alongside.
What is deliberately absent: .env, node_modules, and the live service
definitions. The .example files next to them say what belongs there.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Lucio Lelii