# Run-Record Sheet

Use it for each run of your workflow on one test case. Fill in a new sheet for the first run and another for the re-run after a fix, so the two can be compared.

## The run

- Date:
- Workflow name and version:
- Tool and version (for example, the model name):
- Test case (from your Test-Set Sheet):
- Which run is this? (1 = first, 2 = after a fix, and so on):

## What each check found

Mark each check Met, Not met, or For my read. Quote the line that shows it.

| Check | Mark | What you found (quote the line) |
| ----- | ---- | ------------------------------- |
|       |      |                                 |
|       |      |                                 |
|       |      |                                 |
|       |      |                                 |
|       |      |                                 |

## Score

- Checks met:
- Checks that count (leave out the ones left for your read):
- Checks left for my read:

## What failed, and the fix

Skip this on a run where every check was met.

- Which check failed:
- The one thing I changed (the prompt, the hand-off or the check):
- Why that change should help:

## After the fix

- Same test case, same checks: did the failed check pass now?
- The other test cases, run again: did the fix break any of them?
- Checks met after the fix, out of how many:

## Notes

Anything a later reader needs: a tool that changed, a criterion you reworded, a case you added.

---

From Logic Lab (logiclabhq.com). Free to use and adapt, with credit.