owner
Record the reason
Every change connects to a problem, metric, complaint, task, or roadmap note.
Do not change copy, pricing, routes, or email only because it sounds prettier today.
verification rituals
Operating checklists
A unified set of checks before deploy, page publishing, customer replies, email changes, or experiments.
A checklist does not make the team slower. It keeps a small mistake from putting on a crown and becoming an incident.
owner
Keep receipts
After important changes leave a short changelog: what changed, where, how to verify, how to rollback.
Do not leave the future team facing a riddle without a map and lamp.
everyone
Protect the user
If the user is vulnerable, anxious, or asks for risky advice, move the response back into safe boundaries.
Do not promise healing, guaranteed futures, legal wins, money, or influence over another person.
Deploy checklist
Before publishing a technical change.
- typecheck, lint, test, links, route audit, structure, security.
- Known build caveat documented if container build times out.
- Rollback path exists and owner knows it.
Public content checklist
Before publishing a page visible to users and search engines.
- No forbidden claims, no fear pressure, no medical/legal/financial advice.
- Title, description, internal links, and next step are clear.
- Tone is OmenHall, not generic mystic fog.
Internal page checklist
Before adding internal sections.
- robots noindex/nofollow and X-Robots-Tag are present.
- No customer personal data appears in examples.
- Route is in structure and allowed in route audit.
checklist
- Every check has a yes/no outcome.
- A checklist cannot be closed with 'seems fine'.
- If a check was not run, it is explicitly recorded.
handoff
- After deploy hand owner changed routes, checks run, known caveats, rollback path.
- After content update hand editor pages touched, SEO note, forbidden-claims check.
red flags
- Change without a list of affected routes.
- Internal page without noindex.
- Experiment without rollback rule.
related doors