Runbook de incidentes y rollback

[Kalebtron1]

Why this matters

A single runbook reduces incident time and lowers the chance of ad-hoc fixes.

Problem

There is no document that explains what to do when a deploy or flow fails.

Scope

• Add a short incident runbook with rollback steps and contact points.
• Document the production URLs, env vars, and deploy verification steps.
• Keep the runbook short enough to follow during an incident.

Implementation guidance

• Use docs/flows.md, README.md, vercel.json, and .env.example as the main references.
• Document the normal deploy path and the rollback path.
• Link the runbook from the main README if useful.

Out of scope

• Building a full support process or on-call rotation.
• Writing a long-form architecture guide.

Acceptance criteria

• PR includes the runbook content and links to the relevant files.
• Reviewer can follow the document to reproduce the deploy or rollback steps.
• The PR explains what checks should be done after a deploy.

Validation

• Review the runbook against the current deploy flow.
• Confirm every command or path in the document exists in the repo.
• Run the deploy path locally or document why it is not runnable locally.

PR requirements

• Include Closes #37.
• Keep the runbook concise and action-oriented.
• Note any production-only steps that cannot be reproduced locally.

Complexity

• Trivial (50 points)

[Jujukoder]

@Jujukoder has applied to work on this issue as part of the Stellar Wave Program's 5th wave.

Good Morning, Maintainer. How are you doing? I'm interested in this issue, and I can solve it.

Thank you in advance!

Please Again!

ℹ️ Repo Maintainers: To accept this application, review their application or assign @Jujukoder to this issue.