Inventory the boundaries the demo skips
List every external service, stored record and privileged operation. Exercise the app as a second user and with an invalid or repeated request. Record what happens when a dependency is unavailable rather than assuming the happy path generalises.
Identify missing product decisions: who owns a record, how refunds work or which account can delete data. These are requirements the code generator cannot settle by choosing a convenient default.
Build a release candidate around evidence
Select a narrow set of critical journeys and protect them with repeatable checks. Verify deployment configuration in a separate environment with synthetic data. Keep secrets and production access outside prompts and test fixtures.
Tie the candidate to a commit, the tested configuration and the artifact you plan to release. A test of one working copy does not prove that another build or deployment contains the same result.
Rehearse failure and recovery
Write a rollback procedure and verify it where it cannot harm live data. Consider whether a schema change is compatible with the previous application version. A redeploy is not always a data rollback.
Keep the first release scope small. If the gaps require a new identity model, payment architecture or operational team, treat those as separate design work rather than hiding them inside a promise to clean up the prototype.
Working sheet
| Boundary | Evidence needed | Decision when missing |
|---|---|---|
| Identity and ownership | Allowed and denied user fixtures | Define access rules before release |
| Persistent data | Migration and restore rehearsal | Separate data recovery work |
| External callbacks | Duplicate and failure fixtures | Specify idempotency and retries |
| Release artifact | Commit, configuration and rollback note | Create a reproducible candidate |
Questions about from vibe-coded prototype to production
Can a prototype go live after one cleanup sprint?
Possibly, if its gaps fit the agreed scope and its release conditions are met. There is no general guarantee for an arbitrary prototype.
Does a successful deployment mean production readiness?
It proves that a deployment ran. It does not establish access control, recovery, monitoring or the correctness of the application’s business rules.
How should I use the working sheet?
Download the CSV and replace the sample rows with evidence from your repository. Record a clear acceptance check and an owner for each item. Keep unresolved assumptions visible rather than treating the example as a completed assessment.
Sources and further reading
Examples are illustrative. They are not customer results, security guarantees, or a substitute for checking the documentation for your installed version.
