BATCHPROOF COMPATIBILITY TEST KIT
Version 1.0 | Fixture date 2026-10-25

ALL NAMES AND ORGANIZATIONS ARE FICTIONAL.
This kit is a software test. No shop, production-software or physical engraving result has been established. All operator measurements start NOT RUN.

START
1. Open index.html in a browser for instructions and a printable worksheet.
2. Open the live BatchProof workspace separately. Use Open job to load 01-valid-reviewed.batchproof.json. It contains 12 records, zero blocking errors and 3 warned records.
3. Compare 01-matching-proof.pdf with the 12 individual files inside 01-artwork.zip. Their proof ID must match: BP2-9cd6b2fc-cf555226.
4. Import dimension-check.svg into the shop's actual production software. Record product/version and importer options. Measure the OUTER EDGES of the marked 25 x 10 mm rectangle and 10 x 10 mm square. Do not measure the whole SVG selection.
5. Import a normal, accented and long-name plate; check every field, scale, counters and placement. The SVG page is 76.2 x 25.4 mm. Some software selects ink bounds instead of the page; do not resize the whole selection to those page dimensions. Verify positioning/jig alignment separately.
6. Only an experienced operator chooses machine/material settings and runs a scrap-material test. This kit contains no cutting, power, speed or toolpath instructions. Record actual dimensions, the shop's tolerance and pass/fail; do not assume a universal manufacturing tolerance.

EXPECTED APPLICATION RESULTS
Valid fixture: 12 rows. Rows 1 and 12 intentionally repeat Jordan Ellis with different awards; both get duplicate-name warnings. Row 5 gets an auto-fit warning. Warnings require human review but are not evidence of a spelling error.
The built-in internal review was created only to exercise export. It is labeled QA fixture — no customer approval. Do not represent it as customer consent or production certification.

Open 02-blocking-cases.batchproof.json in a separate test job. Each of its 5 records should have a blocking issue; artwork export must stay disabled. 02-draft-proof.pdf is for visual review only.
- Row 1: missing-name.
- Row 2: unsupported-character (trophy emoji).
- Row 3: invisible-character (U+200B between Alex and Lee).
- Row 4: multiline-field (line break between Jane and Doe).
- Row 5: text-overflow (45 letter Ws).

Open 03-stale-review.batchproof.json. Row 1's year changed to 2027 after the fixture review. Expected current proof ID: BP2-7c1eb2f9-9f52e19e. The retained old review must be stale and batch artwork disabled. Do not record a new review until the revised proof is checked.

FILES
01-valid-roster.csv: UTF-8 CSV for import tests; one row equals one plate.
01-valid-reviewed.batchproof.json: reopenable fictional job with test-only review.
01-matching-proof.pdf: actual BatchProof output, marked SAMPLE.
01-artwork.zip: 12 outlined SVGs, manifest, recipient register and readme.
02-invalid-roster.csv: separate invalid-input CSV.
02-blocking-cases.batchproof.json: reopenable invalid test job, no review.
02-draft-proof.pdf: actual draft output; never production artwork.
03-stale-review.batchproof.json: exact revised job with stale review.
dimension-check.svg: 50 x 35 mm page, marked 25 x 10 mm and 10 x 10 mm outer-edge frames.
measurement-sheet.csv: expected targets plus blank observations; NOT RUN is not pass.
expected-results.json: machine-readable expectations and artifact SHA-256 digests.
index.html: operator instructions and printable measurement worksheet.

INTERPRETING RESULTS
Compare the roster, proof and artwork, not just the filename. Verify imported paths are filled lettering, not a single-stroke font. Check letter counters, accents, apostrophes and long names. Run a test plate before a batch. Any wrong/missing name, silent substitution, clipping, revision mismatch or dimensional result outside the shop's chosen tolerance blocks acceptance.
A passing run applies only to the recorded software version, import options, machine, material and setup. It is not universal compatibility certification.

PRIVACY AND RECORDS
Use fictional or shop-owned de-identified data. The files contain no customer results. The HTML worksheet does not upload or save measurements; print it or record results in the CSV. Keep completed test evidence with the shop.
Proof IDs are version checksums, not verified signatures. File SHA-256 digests check downloaded bytes, not identity or approval. Fixture content and proof IDs are reproducible; export build timestamps may change when the kit is rebuilt.