# Safe to run twice?

Prompt · Correctness and reliability · R01

Finds actions that do damage when they run twice: double sends, double charges, duplicate rows.

## Prompt

```text
Find operations that cannot safely be retried. Examine duplicate events, partial failures, idempotency, transaction boundaries, and recovery from interrupted execution.
```

**What it returns:** Each operation with its trigger, what a second run does, and whether a key or constraint prevents it.

**Skip it when:** Actions that only read data.

**Tools:** Repository and read-only access to the database layout.

**Changes:** Changes nothing

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

## Skills

- [diagnosing-bugs](https://www.atlasscope.dev/skills/diagnosing-bugs): Diagnosis loop for hard bugs and performance regressions.

## Related

- [Updates that overwrite each other](https://www.atlasscope.dev/prompts/updates-that-overwrite-each-other): Finds updates that overwrite newer data because two things wrote at once.
- [Half-finished failures](https://www.atlasscope.dev/prompts/half-finished-failures): Finds what is left behind when something fails halfway.
- [Imports that fail quietly](https://www.atlasscope.dev/prompts/imports-that-fail-quietly): Finds rows that go missing or land on the wrong record during import, with no error.
- [Can we rerun everything without damaging the data?](https://www.atlasscope.dev/prompts/can-we-rerun-everything-without-damaging-the-data): Imports get retried, schedulers overlap, files are replayed, or runs fail halfway.

Source: https://www.atlasscope.dev/prompts/safe-to-run-twice