Image
MVP to Stable

Hardening a vibe-coded MVP without a rewrite

The product has users, and every change breaks something nobody expected. This is the repair-in-place path: pin what the paths users depend on do today, then propose one small change at a time, each read and merged or sent back by a named reviewer on your side.

The short answer

Short answer

You make a working AI-built MVP stable by repairing it in place.

First pin what the paths your users depend on do today with tests, then change the code in small proposed diffs, one at a time. A named reviewer on your side merges or sends back each diff, and the product keeps serving users throughout.

Today

Why does every change seem to break something?

Your product has users and earns its keep. Each new change breaks something nobody expected, and nobody can say which parts are safe to touch, because the code grew faster than anyone's grasp of it. Starting over looks tempting. Meanwhile the product still has to serve those users every day, and a rebuild has to reproduce every behaviour they rely on, including the ones nobody ever wrote down.

Who is repair in place for?

A founder or a small team whose product was built mostly with AI, works, has users, and now has to survive the next round of changes. You may already hold a list of what is wrong from the assessment checklist page in this series, or you are about to get one. What you want next is the product fixed where it stands, with someone on your side approving each change.

What does repair in place touch, and what does it leave alone?

The work is confined to the code you already have, in pieces your owner put on the list. Everything else keeps running as it is.

Part of the productBefore anything changesWhat the agent proposesWho decides
A path your users depend onIts current behaviour is pinned by testsA change only once the pinning tests pass on the current codeYour reviewer, diff by diff
A part named in an accepted findingOn the owner's list, in the owner's orderOne small diff with its reason, the finding it closes and its test resultsOwner sets the order; reviewer merges or sends back
Secrets sitting in the codeRecorded where they sitA diff that moves them out of the codeYour reviewer
Working code no finding namesLeft exactly as it isNothingYour owner, by adding a finding
A part repair cannot make safeEvidence gathered on whyA written flag and no diffYour owner, part by part
Merging into your main codeHeldNever merges its own workA named reviewer, every time

How it works, step by step

01

Start from the accepted findings

The agent works only on what your owner accepted from the assessment, in the order the owner set. Anything outside that list stays untouched, however tempting a clean-up looks.
02

Pin current behaviour first

Before a path your users depend on is changed, the agent adds tests that record what it does today, right or wrong. Michael Feathers describes the purpose of characterization testing as documenting what a system actually does, as opposed to checking for the behaviour you wish it had. A change that breaks a pinned path then fails a check before it reaches anyone.
03

Propose one small change at a time

Each change arrives as a single proposed diff with its reason, the finding it closes and its test results, small enough for a reviewer to read in one sitting. Where a change restructures code, it is refactoring in Martin Fowler's sense: altering the code's internal structure without changing its external behaviour.
04

Leave the rest running

Code that no finding names stays as it is. The product keeps serving its users throughout, because each diff lands on its own and nothing waits for a cutover.
05

Human gate: your reviewer merges or sends back

A named reviewer on your side reads each proposed diff with its test results, then merges it or sends it back with a reason. The agent never merges its own work, and nothing reaches production without that person, whose access stays theirs. The merge step is held the way How It Works describes, and the page on placing approval gates explains how to choose where a sign-off like this one belongs.
06

Record why each change was made

Each merged change keeps its reason, the finding it closed and the name of whoever approved it. The next engineer starts from that record instead of a guess.
07

Flag what repair cannot fix

Where a part cannot be made safe in place, the agent says so and attaches the evidence. Whether to rebuild that part is your owner's decision.

Limits

What limits does repair in place keep?

Proposals only

Every diff goes through code review by a person on your side, who merges it, with no exception for small or urgent diffs.

The accepted list is the scope. Scope grows only when your owner adds a finding to it.

No production access

The agent holds no production credentials, and access to the repository is whatever you grant.

No promise about every part

Where a part cannot be saved, the evidence goes to your owner, who decides.

Where this fits

MVP to Stable

Repair in place is the change path through Mercury's MVP-to-Stable service, meant for a product that works, has users and now needs to hold up under change. It picks up where the assessment checklist stops. Parts of the product that live outside your code, on systems nobody documented, are handled on a sibling page. Once the product holds, the question turns to which work around it agents could run. The use-case library covers that, starting with its readiness checklist for a workflow.

Repair in place or rewrite: how do the two compare?

A rewrite can be the right call for one part, and this comparison does not argue otherwise. Where a part does have to be replaced, that can happen gradually too: Strangler Fig is the name Martin Fowler gave to a gradual approach to legacy modernization.

QuestionRewrite from scratchRepair in place
What users run meanwhileThe existing product, while the new one is built alongsideThe existing product, improving one reviewed diff at a time
Where current behaviour is recordedRediscovered and rebuiltPinned by tests before each change
Unit of reviewThe new system, at cutoverOne small diff
Fits whenYour owner decides, on evidence, that a part cannot be made safe where it standsThe paths users depend on can be pinned and changed piece by piece
  • Approval gates: how to decide which step needs a named person's sign-off.
  • Proactive drafting: another case where the agent drafts and nothing takes effect until a person promotes it.
  • Plans vs routines: why a finite job like hardening is run differently from a recurring process.
MVP to Stable Product
How It Works

Questions, answered

It can say so. The agent flags that part with the evidence, and your owner decides, part by part. Repair in place is where the work starts, and the evidence decides each exception.
AI Opportunity Audit · MVP hardening

Start with the part you are most afraid to change

Bring the part of the product you are most afraid to change to a scoped product review.