Inside the report

A finding should show its work.

Start with the passages. Understand the interpretation. Decide what to do next.

This is a format illustration. The two test entries below were written to demonstrate the presentation. This page is not generated audit output, a customer result, or evidence of measured accuracy.

01 / Evidence and interpretation

Check the journey’s duration.

TimelineIllustrative finding

The two entries give different elapsed times. Whether that is a continuity problem depends on whether they describe the same journey in the same time frame.

Invented test entry A · chapter 4
Departure: 08:00.
Arrival: 10:00.
Invented test entry B · chapter 9
Elapsed travel time: six hours.
Observation
Entry A implies two hours if the times share a date and time zone; entry B explicitly states six.
Possible impact
A conflicting duration may change the sequence of nearby events. Its importance depends on the surrounding story.
Confidence
The arithmetic is straightforward. The relationship between the entries still requires context.
Check first
Journey identity, date, time zone, any intermediate stop, and whether a narrator or in-world record is meant to be unreliable.

02 / A revision decision

Give the author a useful next step.

Locate both passages and check the surrounding text. If they describe the same elapsed journey, reconcile the timing. If they describe different stages, clarify that distinction only if a reader needs it.

Illustrative revision-workbook row
Where Action Author decision
Chapters 4 + 9 Compare journey records and nearby events. Pending context review

You might revise, retain the passages with a reason, or mark the issue for a human editor. The report should support that choice.

03 / Scope and limits

Say what was actually checked.

This illustration contains only two invented entries. It supports no claim about a manuscript, the system’s recall, or false-positive rates.

In a delivered audit, the coverage section distinguishes completed checks, unavailable checks, and candidates that were not individually adjudicated. An unreviewed candidate is not a confirmed defect.

A quotation appearing in the source does not, by itself, establish that the surrounding editorial interpretation is correct.
Read about the evidence checks

04 / After you revise

Recheck the evidence.

The included second full read applies to the revised version of the same work. Supported findings can be compared against the earlier version when the baseline is available.

A finding can change, persist, or need another look. If an earlier finding is not surfaced again, that absence alone is not proof that it has been fixed.