SEO crawlers were downloading the release archives (up to about 2 GB each)
even though they are served noindex. Disallow only *.tar.zst and *.zip under
/tarballz/; the index page and the PKGBUILDs stay crawlable. Repeated in every
group, because a crawler with its own group ignores User-agent: *.
Claude-Session: https://claude.ai/code/session_01GGe7kUhooMFuzY6GaDb2w9
Replaces the placeholder repo with the asger.weirsoe.dk site source.
Adds 6 blog posts (Proxmox, Authentik/OIDC, NPMplus migration, JS injection,
ah-fail) and .gitea/workflows/deploy.yml (bun build + rsync over SSH).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>