Answers the page: what fired, what changed, what the runbook says, and what to tell the channel.
Built for: The engineer holding the pager who has to be useful in the first ten minutes, at two in the morning.
The button opens the checkout, where 6 servers and 4 skills are listed one by one with what each does to the bill — free, already yours, monthly or a one-off licence. Nothing is charged until you confirm it there, in Stripe’s own card frame on that page rather than a redirect, and each paid member keeps its own budget cap.
The reason to buy a stack rather than five listings: each line below needs more than one member connected at the same time.
Read the alert and the metric behind it in Datadog or Prometheus, and say what the series looked like before it fired.
Inspect pods, deployments and services on the cluster and name the one that restarted.
Search LogScale for the window around the alert instead of scrolling a console by hand.
Post the state of play into the Discord channel that is already asking, in the format your team writes updates in.
What each part contributes, and why it was picked over the obvious alternative. Prices and permissions are read from the listings, so nothing here can disagree with the catalogue.
The Datadog API where the alert and its dashboards live, so the first question — what is on fire — is answered from the source.
Queries the series directly, which is what you need when the dashboard shows a spike and you want the raw numbers behind it.
Inspects and manages pods, deployments and services — the layer where most pages turn out to have started.
Ties the page to the release: errors, events and the deploy that preceded them, which is the fastest correlation available in an incident.
The channel the incident is already being discussed in, so the update lands where people are looking rather than in a document nobody opens.
Queries CrowdStrike LogScale for the window around the alert, which beats scrolling a log console at two in the morning.
Keeps the runbook in one shape — service, alert table, dashboards, copy-pasteable procedures — so the page is answered from a document rather than from memory.
Writes the incident update in the format your company already uses, which is what stops the status post from becoming its own argument.
The other half of on call: after the incident, the structured logging and metrics that would have made this page shorter.
Keeps the first ten minutes systematic — evidence, hypothesis, test — instead of restarting things until the graph looks better.
Every member is a subscription or a licence bought once, so the monthly figure is a price rather than an estimate: what moves it is adding or dropping a member, not how hard the stack is worked. The one assumption is that a one-off licence is spread over a year so it can sit in the same column as a subscription.
| Part | What you are paying for | Monthly, as quoted |
|---|---|---|
| Datadog | Free | — |
| Prometheus | Free | — |
| Kubernetes | Free | — |
| Sentry | Free | — |
| Discord | Free | — |
| Logscale | from $25/mo | $25/mo |
| Skills | ||
| Engineering Runbook | $89 · $7.42/mo over 12 months | $7.42/mo |
| Internal Comms | Free · context cost only | — |
| Python Observability | $19 · $1.58/mo over 12 months | $1.58/mo |
| Debugging and Error Recovery | $19 · $1.58/mo over 12 months | $1.58/mo |
| Everything above | $25 of servers plus $11 of skills, the same in a quiet month and a busy one | $36/mo |
Subscriptions at their monthly plan price; one-off licences spread over 12 months. One-off purchases in this stack total $127 — Engineering Runbook $89, Python Observability $19, Debugging and Error Recovery $19 — paid once and spread here so they sit in the same column as a subscription. Everything arrives on one mcprush invoice, taken by Stripe from the card on your account, not one per publisher — mcprush.com is the merchant of record and each publisher is paid out of it.
Install only these and the bill is nothing: 5 servers and 1 skill, 0 of the 2 tools.
Left out, and what goes with it:
Logscale at from $25/mo. Each of those plans states the calls it includes in a month, and running past one never arrives as a larger invoice: the gateway refuses the call over the allowance and returns an MCP error naming the plan. The figure above is what the stack costs in a busy month as well as a quiet one — what a heavy month changes is which plan you need, not what this one bills.
Budget caps are set per install and enforced at the gateway, so a retry loop is refused at the cap rather than left to run through an allowance overnight.
One step per part, in the order they are useful: connect what the work reads before what it writes, and install the skills that decide how the work is done last. Each step is a command you can read before you run it.
npx mcprush@latest stack add on-callNothing in this stack installs from one command today: 10 members are either paid, run from its own source, or a skill with its own command — the steps above name each one. Nothing is connected until you approve it.
Installed together, these tool surfaces add up. It is the first thing a security reviewer asks for, so what has been counted — and what nobody has counted yet — is on the page rather than in a PDF.
| Member | Tool surface | Write tools |
|---|---|---|
| Datadog | not imported | not recorded |
| Discord | not imported | not recorded |
| Kubernetes | not imported | not recorded |
| Prometheus | not imported | not recorded |
| Sentry | not imported | not recorded |
A stack's blast radius is the union of its members, not the worst of them. That union cannot be taken here, because 5 members of 6 have no imported tool surface — so the figure a review asks for is missing rather than low, and a member marked not imported is one nobody has counted rather than one that cannot write.
A stack is a default, not a verdict. These are the substitutions the maintainer would make, and what each one costs or saves.
A shop whose dashboards and logs are in Grafana queries them directly, and the desk stops depending on a paid monitoring account.
The same update, into the workspace where your team actually gets paged, over a user-token Slack server.
Written by the maintainer, kept on the page rather than in a support thread.
It cannot roll back or scale anything on its own account. Kubernetes here inspects and manages what you grant it, and a rollback is a decision the human on call makes.
It does not page anyone. Routing, escalation and acknowledgement stay in the tool that owns your rotation.
Metrics answer what happened, never why. The runbook and the error tracker are in this stack because the graph alone has never closed an incident.
A stack has an owner: whoever keeps it re-checks the combination when a member changes, and the members themselves are published by the people named on each row.
Takes a green branch to a running deployment, and keeps the DNS, the cluster and the access list honest on the way.
Reads the ticket, finds whether it is a bug or a billing question, and drafts the reply in your own voice.
Reads the branch, runs it in a throwaway sandbox, and leaves the review on the merge request before a human opens it.