Build something that is not there yet
Tie the general ledger to its subledgers and classify every break
- When month-end stalls every period because the GL doesn't tie to the AR and AP subledgers.
- When the tie-out is a spreadsheet exercise and nobody can say why last period's break happened.
Build a tie-out model between the general ledger and its subledgers at account, legal entity, and fiscal period, with every break classified by mechanism and an unreconciled-balance metric to trend. Covers reconciliation and break classification, not the correcting journal entries themselves.
Sample This Recipe has not been materialized in the Cookbook repository yet. Its trigger, description, prompt, agent guidance and acceptance conditions, and the explanation below, are prototype drafts. Its name, job, area and readiness come from the reconciled Cookbook seed snapshot. Readiness is a separate question from this one: it says whether the capability exists, not whether the writing has been reviewed.
Use this Recipe
Use the VibeData Recipe `finance-gl-subledger-tieout` at https://getvibedata.ai/cookbook/finance-gl-subledger-tieout Read the Recipe and execute it in the context of the current Intent.
Recipe id finance-gl-subledger-tieout · Not yet materialized in the Cookbook repository, so the pointer addresses this page.
Verified by
What has to be observably true before this Recipe is finished.
- every control account ties within the agreed tolerance for a fully posted period
- a seeded untied period fails the test instead of passing silently
- each break carries exactly one classification (timing, unposted batch, fx revaluation, mapping gap, manual journal)
- breaks are never netted against each other to force a period to tie
Recipe promptThe task specification the agent reads. Reference only — it is not what you copy.
Deliver: Tie the general ledger to its subledgers and classify every break. Build a tie-out model between the general ledger and its subledgers at account, legal entity, and fiscal period, with every break classified by mechanism and an unreconciled-balance metric to trend. Covers reconciliation and break classification, not the correcting journal entries themselves. Execute inside the current Intent. Its Domain, repository, platform, environment and attached sources are the context for this work — read them rather than asking for them. The work is done when: - every control account ties within the agreed tolerance for a fully posted period - a seeded untied period fails the test instead of passing silently - each break carries exactly one classification (timing, unposted batch, fx revaluation, mapping gap, manual journal) - breaks are never netted against each other to force a period to tie Report the evidence for each condition above with the result. A condition you cannot meet is something to say, not something to work around.
Agent guidanceHow the agent approaches the work, and what it will not do.
Profile the inputs the grain, joins and measures actually depend on before proposing a model. Put the design up for review — grain first — then build in an isolated copy with tests and documentation landing beside the model rather than after it.
Composes
- dbt model authoring and layering
- dbt in the project
- dlt in the project
- isolated-copy execution and gate verification
Asks first
Semantic decisions the Intent cannot supply. Never context Studio already holds.
- which Fabric target this work lands on, when the Domain carries both a Lakehouse and a Warehouse
- the grain the requester expects, where the request leaves it open to more than one reading
Guardrails
- Build in an isolated copy. Production is read, never written.
What you need
- Microsoft Fabric Lakehouse, Microsoft Fabric Warehouse, MotherDuck, or DuckDB.
- A dbt project you can build, and read access to the models it starts from.
- dbt and dlt in the project, or the intent to add it.
- Needs the subledger-to-control-account mapping supplied up front.
How it goes
- State the outcome in one sentence, in the language the request arrived in.
- Let it profile the inputs the grain, the joins and the measures actually depend on.
- Review the design. Disagreeing about grain here costs a sentence; after the model exists it costs a rewrite.
- Let it build in an isolated copy, with the tests and the documentation landing beside the model rather than after it.
- Read the acceptance conditions against the run.
What you end up with
The deliverable, in your own repository, as transformation work a reviewer who knows the project reads as native to it. Alongside it, the evidence for every one of the acceptance conditions above — which is the part that is still there in three weeks when somebody asks.