From Runbook to Script: Turning My n8n Pre-Publish Checklist Into a One-Command Gate

Description and use

The ./prepublish.sh script replaced a manual five‑minute n8n LLM workflow pre‑publish checklist, enforcing all gates and uncovering hidden failures that the manual process missed.

Problem

A twenty‑minute manual pre‑publish checklist for n8n LLM workflows was frequently skipped, risking untested deployments.

Outcome

The script reduced manual effort, enforced non‑skippable gates, and revealed three hidden failures: unasserted sabotage, nondeterministic crash points, and vacuous cost rehearsal in CI.

Workflow

• Run fixture clean and adversarial checks
• Iterate over providers (openai, anthropic, gemini) forcing failure and verifying injected failure and fallback
• Force crash after node extract_3, then resume and verify no duplicates or re‑spend
• Run cost rehearsal by checking escalation rate <= 0.15 and per‑run cost <= $0.50

Evidence

My last post ended with a five‑step pre‑publish checklist for n8n LLM workflows. It took twenty minutes to run by hand.

Source

I turned it into one command: ./prepublish.sh. Any gate fails, exit non‑zero, nothing ships.

Source

Three things the script caught that the checklist never did…

Source