Treat content withdrawal as a reversible migration
Withdraw published content without destroying its source, revision history, or the evidence needed to move it safely.
Withdraw published content without destroying its source, revision history, or the evidence needed to move it safely.
How this blog moved AI drafting onto a local desktop while keeping capacity checks, validation, deployment, and social delivery behind reproducible cloud gates.
Design a qualification harness that keeps the first failure, accounts for uncertain resources, and publishes useful evidence without exposing secrets.
A stray PHPStan cache in a nested git worktree got scanned as production code, showing why filesystem-exclusion rules for local gates need to match at any depth.
A missing npm ci let schema validation crash on every mined candidate, filing zero content-queue issues for a week — the fix layers a loud fail-open guard over the actual root-cause patch.
How a human-triggered ‘@claude review’ workflow got locked down to read-only, bounded-diff, revision-verified reviews after the official tag-mode example turned out to be implementation-capable by default.
First-run lessons from an automated content pipeline. Noise, human-only merges, and a backwards production step surfaced in 24 hours.
Build a daily content mining pipeline that scans your repos and queues post ideas as GitHub issues.
Automate backups, test restores, and build a disaster recovery runbook for your VPS.
Build a script that generates consistent OG social card images from an HTML template using Playwright screenshots, with smart caching and series-aware gradients.