The DevOps & Reliability Runbook
136 copy-paste AI prompts for platform and infrastructure teams — deployment strategy, on-call that people can sustain, SLOs and error budgets, blameless postmortems, cloud cost and migration plans.
Key Capabilities & Features:
- •The Platform Primer: paste your stack, scale, team size and on-call arrangement once — every prompt after it writes for your environment
- •On-call designed to be sustainable for two years rather than for a quarter, because rotations do not fail on the rota, they fail on attrition
- •Postmortems that produce contributing factors rather than a single root cause, since a system that failed for one reason alone is a system you have not understood
- •Cost prompts that count engineer time against the saving, which is how most optimisation projects quietly lose money
- •Cloud-agnostic — the Primer asks which provider, and the prompts name provider-specific mechanics only where they genuinely differ
- •136 prompts across 8 sections — deployment & release strategy, infrastructure as code review, slos & error budgets, on-call & alerting and more
- •Works with ChatGPT (GPT-5), Claude Opus 5 and Gemini 3.7 Flash — fill the bracket, paste, done
One-time payment • Lifetime access & future updates
What's Inside
8 modules136 promptsEvery module below is included in the single purchase.
- 1
Deployment & Release Strategy
20 promptsChoosing between rolling, blue-green and canary on the properties of your system rather than on fashion, rollback that has actually been tested, and the release nobody can undo.
- 2
Infrastructure As Code Review
18 promptsReviewing a change for blast radius before it merges, state and drift, the module that is one resource away from being unusable, and secrets that should not be where they are.
- 3
SLOs & Error Budgets
18 promptsObjectives set from what users notice rather than from what is easy to measure, error budgets people honour, and the alert threshold that was chosen because it was round.
- 4
On-Call & Alerting
20 promptsA rotation people can sustain for two years, alerts that mean act rather than notice, silencing the pager honestly, and the handover note the next person needs at 3am.
- 5
Incident Communication
18 promptsThe status page update during an outage, telling customers before they tell you, internal coordination roles, and communicating an incident you do not yet understand.
- 6
Postmortems
18 promptsBlameless in practice rather than in the title, contributing factors instead of a root cause, action items with owners, and the postmortem that produces no change.
- 7
Cost & Capacity
14 promptsReading a cloud bill for what is actually driving it, the environment nobody switched off, capacity planning against real growth, and the saving that costs more in engineer time.
- 8
Migrations
10 promptsSequencing a migration so each step is independently reversible, the cutover plan, running two systems in parallel, and knowing when a migration should be abandoned.
Test Workflow Execution in Your Browser
Verify actual prompt structure, token speed, and reasoning depth before downloading.
Frequently Bought Together
Technical Specifications
Architecture & System Overview
Most outages are not caused by the thing the postmortem blames. They are caused by three conditions that were each individually tolerable, and by a runbook nobody had opened since it was written.
The DevOps & Reliability Runbook is a 136-prompt playbook covering deployment and release strategy, infrastructure as code review, SLOs and error budgets, on-call and alerting, incident communication, postmortems, cost and capacity, and migrations.
Start with the Platform Primer — paste in your stack, scale, team size and on-call arrangement — and every prompt after it writes for your environment rather than for a company with fifty engineers.
The on-call section is designed around sustainability. A rotation is easy to design and hard to survive, and the failure mode is not a bad rota — it is the engineer who leaves after eighteen months of interrupted sleep. The prompts work through page volume per shift, what fraction of alerts are actionable, and the uncomfortable question of whether your rotation is currently viable at your headcount.
Postmortems are blameless in mechanism rather than in name. The prompts produce contributing factors rather than a root cause, because a system that failed for exactly one reason is a system you have not finished investigating. They also test each action item against whether anyone will actually do it — a postmortem generating eleven improvements produces zero.
And the cost section counts engineer time against the saving. A great many cloud optimisation projects save four thousand a year and consume six thousand in salary getting there.
Scope note: runbooks, code review, architecture decision records and technical writing live in The Software Engineering System. This one covers operating the system rather than building it.
Compare with Similar AI Templates
| Specification | ||||
|---|---|---|---|---|
| Customer Rating | 4.88(8) | 4.91(43) | 4.93(45) | 4.91(38) |
| AI Models | GPT-5, Claude Opus 5, Gemini 3.7 Flash | GPT-5, Claude Opus 5, Gemini 3.7 Flash | GPT-5, Claude Opus 5, Gemini 3.7 Flash | GPT-5, Claude Opus 5, Gemini 3.7 Flash |
| Formats Included | PDF Prompt Playbook | PDF Prompt Playbook | PDF Prompt Playbook | PDF Prompt Playbook |
| Lifetime Updates | Yes | Yes | Yes | Yes |
| Creator Rating | Wren Ashcombe (98.9%) | Dr. Lucas Vance (99.2%) | Marcus Sterling (99.3%) | Elena Rostova (99%) |
Community Reviews & Ratings
Verified buyer ratings & prompt benchmarks
Write a Review
Sign in to leave a review for this system.
Verified User Feedback
We had an alerting problem, not an incident problem
About 70% of our pages were acknowledge-and-sleep. Cutting those took two weeks and did more for the rotation than the extra headcount we had been asking for and not getting.
It said our rotation was not survivable
Four people covering nights with no compensation. It gave me the attrition-risk case with replacement cost attached, which is the framing that finally landed with our director.
Contributing factors, not a root cause
Our postmortems had always ended with one person's change. Three factors is both more honest and produces fixes that actually prevent recurrence.
Operating, not building
No runbooks, no architecture records, no code review — it points at the software engineering toolkit for those. Clean scoping, but I did end up buying both.
Frequently Asked Questions
Which cloud is it for?
Any, and on-premises too. The Primer asks, and provider-specific mechanics are named only where they genuinely differ. Deployment strategy, SLOs, on-call design, postmortems and migration sequencing are not provider-specific at all.
Is this exam preparation for a cloud certification?
No. It contains no exam material and is not affiliated with any certification body or cloud provider. It is about operating systems in production, which certifications test indirectly at best.
We are a three-person team. Is this written for large organisations?
The Primer asks your size, and the prompts are explicitly instructed not to propose practices that need more people than you have. Where something genuinely requires a larger team — a follow-the-sun rotation, for instance — they say so rather than shrinking it into a version that does not work.
Does it write my infrastructure code?
No. It reviews a change you describe for blast radius, drift and secret handling, and helps you plan and sequence. The code and the terraform stay with you, as does the decision to apply anything.
How is it delivered?
You receive a PDF with all 136 prompts organized into 8 sections, delivered within 24 hours of purchase. No app or account required — copy a prompt, fill in the bracketed variables, and paste it into ChatGPT, Claude, or Gemini.
Can I use this for client work?
Yes. Your purchase includes a single-user commercial and personal license, so you can use every prompt in your own business or for client deliverables without restriction.