notplantsandClaude Fable 5.1
31af820079
The orchestrator's flake now builds the machine it shares with the cc-ci CI
server: `nixosConfigurations.cc-ci` composes cc-ci's nixosModules.cc-ci-server
(new flake input, nixpkgs + sops-nix follow ours), this repo's orchestrator
module (nix/modules/cc-ci.nix, exported as cc-ci-orchestrator, `cc-ci` kept
as an alias for notplants-nix) and the new nix/modules/orchestrator-host.nix
— the host contract those units always assumed (loops user, claude/opencode
CLIs, opencode web server + tailnet-only UI on 8443 since traefik owns
80/443, nix-ld, tool set, `ssh cc-ci` → loopback).
nix/hosts/cc-ci/{hardware,networking}.nix are PROVISIONAL copies of the old
server's layout so the flake evaluates; they get replaced by the
nixos-infect output of 195.201.88.249.
README.md is the deploy guide (Hetzner Debian → nixos-infect → this flake →
staging → data restore → cutover). archive/ holds the retired Incus/Hetzner
orchestrator host configs, the old terraform and the migration plans;
references updated. cc-ci-plan/plan-cc-ci-combined-host.md is the working
plan for the move.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FqkQq3CDmFWcQ7u1LzoyRz
2.9 KiB
Plan stub — repo consolidation + recipe-maintainer submodule
Status: COMPLETE (2026-05-31). Both cleanups executed. See commit history for details.
Task 1 — merge the two orchestrator repos into one cc-ci-orchestrator
Today there are two repos:
recipe-maintainers/cc-ci-autonomous-orchestrator— the loop code/plans/prompts (cc-ci-plan/,launch.sh,prompts/,references/,AGENTS.md). This is the workspace at/srv/cc-ci(/home/loops/cc-cion the VM).recipe-maintainers/cc-ci-orchestrator— the VM's NixOS config (configuration.nix,README.md), created during the VM migration.
Goal: combine into a single repo named cc-ci-orchestrator (loop code/plans + the VM's NixOS
config in one place). Likely: move the NixOS config into the autonomous-orchestrator repo (e.g.
nixos/ or vm/), then rename that repo to cc-ci-orchestrator on Gitea.
Touch-ups when doing it: the workspace clone URL in launch-orchestrator.sh / kickoff docs /
cc-ci-loops.service, the VM's git remote, and plan.md/kickoff.md references to the repo name.
Keep recipe-maintainers/cc-ci (the CI product repo the loops build) separate — this is only
about the two orchestrator repos.
Task 2 — make references/recipe-maintainer a git submodule
Today references/recipe-maintainer is a symlink → /srv/recipe-maintainer/ (an absolute path).
On the Pi that target existed; on the VM it does NOT, so the symlink dangles and the parity corpus
is empty — Phase-2 parity-porting (P2) reads from references/recipe-maintainer/recipe-info/<recipe>/ tests/*.py, so recipes not already ported can't be ported until this is fixed.
Goal: replace the symlink with a proper git submodule from
https://git.autonomic.zone/notplants/recipe-maintainer:
git rm references/recipe-maintainer # remove the symlink
git submodule add https://git.autonomic.zone/notplants/recipe-maintainer references/recipe-maintainer
git commit
# on each clone / the VM: git submodule update --init --recursive
This both fixes the current VM gap (corpus present + reproducible, no absolute-path dependency) and makes the workspace self-contained on any host.
Interim (before this plan runs), if the loops need the corpus on the VM: clone recipe-maintainer
to /srv/recipe-maintainer on the VM (needs root — /srv is root-owned) so the existing symlink
resolves; OR do Task 2 directly.
Operator action required — Pi clone remote update
The Gitea rename (cc-ci-autonomous-orchestrator → cc-ci-orchestrator) breaks the origin remote
on any existing clone. On the Pi (and any other machine with a clone), run:
cd /srv/cc-ci # or wherever the workspace is cloned
git remote set-url origin https://git.autonomic.zone/recipe-maintainers/cc-ci-orchestrator.git
# (include bot credentials in the URL if needed, as before)
The VM clone's remote was updated automatically as part of this task.