---
name: land-and-deploy-checklist
description: "Walks a merged change through to production safely: merge, wait for CI and deploy, check live health, and roll back at the first bad signal. Use when an approved pull request is ready to merge and the team wants the deploy watched and verified in production with a rollback ready. Do not use for preparing the pull request or running a long post-deploy monitoring window. Related topics: devops, coding."
---

# Land and Deploy Checklist

Walks a merged change through to production safely: merge, wait for CI and deploy, check live health, and roll back at the first bad signal.

## When to use

Use this skill when an approved pull request is ready to merge and the team wants the deploy watched and verified in production with a rollback ready. Do not use it for preparing the pull request or running a long post-deploy monitoring window.

## Workflow

1. Ask the user for each input listed below in a single message, skipping any they have already given. Where options are listed, offer them, but accept other answers too.
2. Confirm approvals, green checks, and an allowed deploy window before merging.
3. Merge with the chosen method and follow CI and the deploy until each reports completion.
4. Verify the served version matches the merge and that key endpoints respond healthy.
5. Compare error rate and latency with the pre-deploy baseline over at least ten minutes.
6. Roll back immediately when a trigger fires, otherwise report a final status of Healthy or Needs attention.

## Inputs

- **Pull request** (free text, e.g. "Link or number")
- **Merge method** (choose one: Squash, Merge commit, Rebase)
- **Deploy pipeline** (free text, e.g. "GitHub Actions to Cloud Run")
- **Health signals** (longer free text)
- **Rollback method** (longer free text)
- **Deploy window** (choose one: Anytime, Business hours only, Off-peak only)

## Output

A timeline of merge, CI, and deploy events, the health checks with results, and a final status of Healthy, Rolled back, or Needs attention.

## Prompt template

Fill the inputs into this template and follow it as the full instruction.

### Land and Deploy Checklist

Take an approved pull request live and confirm it is healthy.

- Pull request: {{Pull request}}
- Merge method: {{Merge method}}
- Deploy pipeline: {{Deploy pipeline}}
- Health signals: {{Health signals}}
- Rollback method: {{Rollback method}}
- Deploy window: {{Deploy window}}

Confirm before merging that approvals are in, checks are green, and the deploy window allows it. Merge with the chosen method, then watch CI and the deploy to completion instead of assuming success. After the deploy, verify production: the version served matches the merge, key endpoints return healthy responses, and error rate and latency sit near their pre-deploy levels over at least ten minutes. Define rollback triggers in advance: error rate above its baseline by a set margin, a failing health endpoint, or a broken core user flow. If any trigger fires, roll back first and investigate second. Report merge time, deploy time, checks passed, any anomaly, and a final status of Healthy, Rolled back, or Needs attention.

## Pitfalls

- Calling the deploy done when the pipeline turns green, without hitting production.
- Investigating before rolling back, which lengthens the outage.
- Using a baseline taken after the deploy, which hides a regression.