PageUtil DMARC Before-and-After Test Pack — version 1
https://pageutil.com/dmarc-report-analyser/

All reports are synthetic. Domains use .example and IP addresses use
documentation ranges. No real customer data, DNS checks or external services
are needed. These are legacy aggregate-format XML fixtures, not a full XML
schema conformance suite. You may reuse and quote these fixtures and expected
findings in evaluations, articles and demonstrations; attribution is appreciated.

HOW TO RUN
1. Extract this ZIP. Do not upload the entire pack as one batch: that would
   mix unrelated cases and reporting periods, and include non-XML instructions.
2. Open the analyser. Clear any previous inputs/results before each case.
3. Select every XML in the chosen case's baseline folder as Baseline reports.
   Select every XML in its comparison folder as Comparison reports.
4. Click Analyse reports. Compare the totals, source changes, coverage cautions
   and expanded report provenance with EXPECTED-FINDINGS.txt.
5. Optionally export both CSVs to retain the underlying evidence.

There are three scenarios; scenario 3 has two independent cases (3a and 3b).
Each folder pair is a separate run. The machine-readable expectations.json
records the same expected counts, source changes and caution fragments used
by the repository's regression test.

Periods use inclusive UTC timestamps. Normal baseline: 2026-01-01 00:00:00
through 23:59:59; comparison: 2026-01-02 00:00:00 through 23:59:59.
Case 3b's comparison stops at 11:59:59 UTC. All policies stay p=none with
relaxed SPF and DKIM alignment to isolate sender and coverage changes.

The sender inventory and configuration fix are fictional background supplied
by this pack, not conclusions proved by XML. No aggregate report can establish
sender ownership, abuse, DNS correctness, inbox placement or causation alone.
No result here is a recommendation to increase DMARC enforcement.

For interpretation and exact expected results, read EXPECTED-FINDINGS.txt.
