Short answer
What you need to know
Choose one concept, read its expected and forbidden findings, download or privately inspect the generated sample, and compare the scanner's observation with the source evidence and rendered reference. Treat the result as evidence to verify, not a conclusion about intent.
Start with the lesson contract
Filter by format, concept, difficulty, or finding type, then open a lesson that matches what you want to learn. Each entry explains what is hidden, why it might be benign, how to verify it, and which findings should and should not appear.
Clean controls show that an ordinary file should remain free of the lesson's forbidden findings and help distinguish a detector signal from a format-wide assumption.
Know which kind of evidence you are viewing
Source evidence identifies stored document content or structure. A rendered reference shows how a supported viewer presents it. Scanner findings organize detected signals. Human conclusions explain meaning or intent and must be made separately.
A rectangle or page reference is not the source itself. Word projections can be approximate, Excel references remain intentionally unannotated and sheet-aware, and PDF geometry should be checked in an independent viewer.
Download versus inspect this sample
Download saves the exact allowlisted teaching file with a fixed safe name. Verify its SHA-256 digest and size against the catalog if you need to preserve its identity.
Inspect this sample first validates the file, then copies it into a normal private temporary scan workspace. The report follows the same owner isolation, deletion, expiry, cancellation, and cleanup rules as an ordinary scan.
Compare expected and observed results
Confirm that every required category appears at the expected source identity, forbidden categories stay absent, and clean controls remain clean. Differences can indicate a scanner change, a renderer limitation, or a lesson that needs review.
Do not force a result to match the lesson. Preserve the observed evidence, reproduce it with the reviewed scanner version, and use the safe remediation exercise on a copy before scanning again.
Use the samples safely
The curriculum is synthetic and declares active content and external relationships explicitly. Downloads fail closed when their manifest, package safety, relationships, size, format, or digest do not match the reviewed contract.
Do not replace the fixtures with an unreviewed file or use the lab as a malware sandbox. For real documents, preserve the original and follow the security and records process appropriate to their source and sensitivity.
Important limitation
Verify important findings
The lab demonstrates selected scanner concepts; it is not a complete document-forensics, malware, accessibility, or authenticity test. Rendering can vary among applications, and expected findings remain subject to reviewed scanner-version updates.
Further reading
Authoritative references
These primary sources support the manual steps in this guide. Their inclusion does not imply endorsement of Hidden Word Scanner.