# Can we independently prove these numbers?

Prompt · Accuracy and evidence · PG01

Important figures may be wrong even though their lineage is documented.

## Prompt

```text
Take a representative set of important figures from this database and independently reconstruct them from their original sources. Don't rely on the existing calculation code to verify itself. Find discrepancies in definitions, source selection, units, dates, classifications, and relationships. Determine which errors are isolated and which come from shared logic, then rank corrections by analytical impact.
```

**What it returns:** Samples from the original sources, an independent rebuild of each figure, each disagreement classified, how many records it affects, and the shared causes.

**Skip it when:** The figures are copied straight from the source with no calculation applied.

**Avoid:** Checking a value with the same code or source reading that produced it.

**Tools:** Needs: the repository, read-only database access and the original source documents.

**Works on:** Any PostgreSQL host: Supabase, Neon, RDS, Cloud SQL or self-hosted.

**Changes:** Changes nothing

**Sessions:** Look first, then fix in a new session

## Related

- [Where did this number come from?](https://www.atlasscope.dev/prompts/where-did-this-number-come-from): Checks that every number on screen can be traced back to a source document.
- [Missing vs zero vs stale](https://www.atlasscope.dev/prompts/missing-vs-zero-vs-stale): Finds where 'we don't know' turns into 0 and then into a wrong total.
- [Find data that is internally consistent but wrong](https://www.atlasscope.dev/prompts/find-data-that-is-internally-consistent-but-wrong): Every check passes, but totals or outputs still look implausible.
- [Reproduce what the database believed in the past](https://www.atlasscope.dev/prompts/reproduce-what-the-database-believed-in-the-past): Past reports, backfills, restatements or changed classification rules matter.

Source: https://www.atlasscope.dev/prompts/can-we-independently-prove-these-numbers