Stage 05 / 10 minutes / Pairs or individual
Plan from the specifications
Produce the smallest ordered plan that covers each accepted claim and its evidence.
- Output
- Implementation plan
- Support
- One claim-to-change example
- Continue when
- All readiness checks pass
Repository for this lesson jhipster/jhipster-sample-app-react at 8c9d248 . All source links, commands and observations refer to this pinned revision.
Orient
Why this stage matters
Explain why plan from the specifications affects confidence in the build.
You can name the decision this stage supports and the risk created by skipping it.
This ties the stage to the lab’s central question: Does the evidence show that posting an operation and changing its account balance form one safe, reviewable outcome?
A plan should expose scope and sequence before code is generated. It should also reveal claims with no implementation step, edits with no accepted purpose and tests that run too late to guide the work.
Inputs
Use repository sources and accepted artifacts
Identify the accepted evidence and artifacts required for this stage.
You can explain why each source is relevant and exclude material that does not affect the work.
These sources establish what the repository does when operation and balance records can be changed independently. Later claims and tests must trace back to this evidence.
Audited specifications
Use only accepted claims, evidence and limitations.
REST resource
Locate the current transaction and API seams.
src/main/java/io/github/jhipster/sample/web/rest/OperationResource.javaAccount repository
Locate account persistence.
src/main/java/io/github/jhipster/sample/repository/BankAccountRepository.javaOperation repository
Locate operation persistence and relationship loading.
src/main/java/io/github/jhipster/sample/repository/OperationRepository.javaBank account resource
Account for direct balance mutation paths when the accepted specification requires them.
src/main/java/io/github/jhipster/sample/web/rest/BankAccountResource.javaReact operation flow
Locate request and result handling.
src/main/webapp/app/entities/operation/operation.reducer.tsWorked example
Map one atomicity claim to a test and code seam
Study one worked example and identify the judgment behind each step.
You can reproduce the reasoning without copying the example’s conclusion.
The example models one part of the operation-and-balance problem. You apply the same reasoning to the remaining paths and produce evidence another engineer can review.
- Select the atomic creation claim and its Java integration test.
- Name the boundary that must own both state changes.
- Order the failing test before the implementation step.
A plan names the purpose and evidence for a change. A file list alone does not show whether the plan covers the specification.
The worked example does not propose the full architecture.
Decide
Choose the repository seams and stop conditions
Separate repository facts from decisions that require human authority.
Each question has an accepted answer or remains blocked with a named owner.
The repository cannot decide credit, debit, retry, ownership or mutation policy. Those decisions define what balance consistency means and what can be judged safe to ship.
- Which boundary should coordinate the operation and balance changes?
- Which test should fail before each behavior changes?
- Does the accepted specification require a schema or response change?
- Which generated files must remain untouched?
- Where should the team stop and revise the specification?
Use the agent
Use the agent on this repository
Use the coding agent to support plan from the specifications while retaining control of decisions and evidence.
You verified each response before continuing and carried only accepted findings into the artifact.
Bounded prompts help investigate and build the balance change while human checks prevent unsupported agent output from entering the evidence chain.
These are task patterns for the pinned JHipster source, not answer scripts. Replace every bracketed field with accepted repository work. Open every cited path at the pinned revision and review each response before continuing. If you cannot supply a field, return to the earlier artifact.
Decide before prompting
- Use only audited claims and test specifications.
- Set the largest change a reviewer can assess in one increment.
Map the work
The first pass exposes missing coverage before ordering implementation.
Map every accepted claim to its test change and smallest code seam. Identify claims with no implementation work and proposed edits with no accepted claim. Do not order the work yet.Trace every row backward to a claim and forward to an observable test result.
Order the increments
Evidence-first ordering shows that existing behavior fails before code changes make it pass.
Order the mapped work into reviewable increments. Put the focused failing test before each behavior change. Give every increment a claim ID, allowed files, verification command and stop condition.Split an increment if it changes more than one independently reviewable behavior.
Add risk and recovery
A plan must show how a failed increment is contained or reversed.
For each increment, state its main risk, evidence limit and rollback or recovery step. Flag generated files and unrelated changes that must remain outside the boundary.Confirm that each recovery step can be performed without guessing.
Audit the plan
A plan should expose specification gaps instead of solving them silently.
Audit the plan for uncovered claims, unclaimed edits, oversized steps, vague commands and decisions made outside the accepted specification. Return findings only.Resolve findings in the plan or return the work to the specification stage.
Create
Produce the implementation plan for this repository
Create a reviewable implementation plan from accepted inputs.
The artifact contains every required field and stays within its stated limit.
This artifact records the evidence or decision the next stage needs to move the operation-and-balance change toward a supported release judgment.
- Create a plan row for each accepted claim.
- Link the claim to a test, implementation seam and verification command.
- Order work in small test-and-behavior increments.
- Record risks, rollback steps and explicit exclusions.
- Check the plan for unlinked edits and uncovered claims.
Six to 10 ordered steps.
- Step
- Claim IDs
- Test change
- Code seam
- Verification
- Risk
- Rollback or recovery
Challenge
Try to disprove the repository-backed work
Find a material weakness before the artifact controls later work.
Each challenge has evidence, an impact and a recorded response.
The challenge tests whether a partial write, duplicate effect, unauthorized operation or unsupported claim could survive the proposed artifact.
Ask another pair to find one uncovered claim, one edit without an accepted purpose or one step that assumes its result. If you are working alone, trace the plan backward from each edit and forward from each claim.
Required responseThe authors update the sequence, narrow the scope or return to the specification.
Revise
Use repository evidence before continuing
Resolve findings and make an explicit decision about readiness.
Every readiness check passes, or the work returns to the section that must change.
The gate prevents unresolved balance-consistency risks from flowing into code or supporting a stronger shipping claim than the evidence allows.
Ready to continue when
- Every plan step links to an accepted claim and evidence.
- Every accepted claim appears in the plan.
- The sequence can produce useful evidence before the full build is complete.
If a check fails
- Remove edits that do not support a claim.
- Split steps that change several behaviors before verification.
- Return to stage 2 or 3 when the plan exposes a missing decision or test.
Transfer
Apply the practice to your work
Map this practice to a real codebase, role and delivery decision in your work.
You can name the equivalent artifact, evidence, owner and next use in your team.
The lab succeeds when you can use the same evidence chain to judge an AI-assisted change in a production codebase, beyond this sample’s operation-and-balance problem.
- Compare this plan structure with your team’s issue or pull request template.
- Name one field you would add to make claim-to-test traceability routine.
Record your answer. The lab is complete only when you can name the equivalent practice, artifact and evidence in your work.