Die öffentlichen Docs sind funktional vollständig und kanonisch, enthalten aber keinen Quellcode, keine privaten Kontodaten und keine reproduzierbaren Tuning-Parameter.
Website — single workspace copy (SSOT for “where is the site?”)
What runs in production (verified)
- Host:
kapitaalbot.nlis served by PM2 processkapitaalbot-website. - Working tree on server:
/srv/kapitaalbot/website(inside theKapitaalBotmonorepo checkout at/srv/kapitaalbot). - Package identity:
package.json→"name": "kapitaalbot-website","version": "1.0.0"(Next 15 stack in this tree).
Conclusion: production HTML/Next routes come from KapitaalBot repo → website/, not from the separate GitHub repo Davelaar/KapitaalBot-Website (that line was Next 14 / different app).
Canonical copy inside this repo (KRAKENBOTMAART)
- Path:
KapitaalBot-Website/at the root of this repository (KRAKENBOTMAART). - Contents: a file-level mirror of
KapitaalBot/website/(the live site source) at consolidation time. No nested.githere — the website is versioned as part of the Krakenbot monorepo. - Goal: one obvious place in the workspace for all website work; no second “mystery” checkout.
Rollback / archives (local)
After consolidation, full backups were written under:
backups/kapitaalbot-monolith-website-YYYY-MM-DD.tar.gz— tarball ofKapitaalBot/website/at consolidation time.backups/kapitaalbot-website-standalone-clone-YYYY-MM-DD.tar.gz— tarball of the old standalone~/KapitaalBot-Websitecheckout.backups/KapitaalBot-Website-standalone-YYYY-MM-DD.bundle—git bundleof that standalone repo (all refs).
Restore standalone repo example:
mkdir -p ~/KapitaalBot-Website-restore
git clone KapitaalBot-Website-standalone-YYYY-MM-DD.bundle ~/KapitaalBot-Website-restore/repo
# or: tar xzf backups/kapitaalbot-website-standalone-clone-YYYY-MM-DD.tar.gz -C ~
Deploy today (unchanged until you migrate)
Server script KapitaalBot/deploy/run_website_deploy_server.sh still does:
WEB="${REPO}/website" # i.e. /srv/kapitaalbot/website
To deploy from this repo’s KapitaalBot-Website/ instead, you must change server layout or that script (Git-only): e.g. checkout KRAKENBOTMAART on the server and set WEB to …/KapitaalBot-Website, or replace /srv/kapitaalbot/website with a symlink after git pull — plan that explicitly; do not mix two trees silently.
Tot die migratie: twee bomen vermijden
- Workspace-SSOT (Cursor): wijzig hier:
KapitaalBot-Website/. - Deploy-SSOT (nu nog):
KapitaalBot/website/in het KapitaalBot-monorepo.
Na wijzigingen in deze repo: synchroniseer bewust naar KapitaalBot/website vóór je KapitaalBot pusht en run_website_deploy_server.sh draait, of migreer de server-build naar alleen KRAKENBOTMAART en pas het deploy-script aan. Zonder een van beide ontstaat drift.
Voorbeeld (lokaal, beide checkouts naast elkaar):
rsync -a --delete \
--exclude 'node_modules' --exclude '.next' --exclude '.env.local' \
/pad/naar/KRAKENBOTMAART/KapitaalBot-Website/ \
/pad/naar/KapitaalBot/website/
Old GitHub repo KapitaalBot-Website
The separate repository remains on GitHub for history/issues if needed; it is not what PM2 ran at the time of consolidation. Prefer issues/PRs against this repo under KapitaalBot-Website/ once deploy points here.