No download or checkout is available.
Evidence before claims
FilePreflight release evidence
A public status page for local-processing, format coverage, conversion quality, and post-check claims.
Pre-release
Methods are public. Release-build results are not yet published.
FilePreflight is still in development. This page deliberately separates a planned test from a passed test. A result will move to “Verified” only when the release candidate and retained synthetic fixture support it.
0release claims marked verified today
Method defined; release-build run not published.
17 extensions planned; pass records not published.
Synthetic PDF example; not FilePreflight output.
Method 01
Local-processing network test
The release test will monitor a clean Windows 11 x64 environment while a supported synthetic file moves through inspection, transformation, reopen, and recheck. The record will distinguish file processing from an explicit Pro purchase or license action.
- Establish the environment.Record the app build, OS, fixture hash, adapters, DNS state, and monitoring tools.
- Run the Free file path.Observe DNS and outbound connections while processing; file bytes and file-derived data must not leave the device.
- Separate licensing.Test an explicit Pro license action independently and confirm that no file data enters the request.
- Publish the boundary.Report conditions, observations, limitations, and artifacts without calling an unobserved condition “zero network.”
Result pending No pass claim is made on this page.
Method 02
Planned launch-format coverage
Free and Pro are planned to support the same 17 extensions with identical per-file inspection, conversion, speed, and post-verification quality. Each extension still needs a reproducible release-build result.
| Group | Extensions | Public status |
|---|---|---|
| Documents and Office | .pdf, .docx, .xlsx, .pptx | Release test pending |
| Images | .jpg, .jpeg, .png | Release test pending |
| Text | .txt, .md, .markdown, .log | Release test pending |
| Data and configuration | .csv, .json, .xml, .yaml, .yml, .env | Release test pending |
A passing record must show that the selected canary is physically absent from the output, useful content remains usable, the output reopens, the second check passes, and the original hash is unchanged.
Method 03
What a public review record should contain
FixtureSynthetic input description, hash, format, and expected finding.
TransformationSelected policy and the category removed—without exposing a usable secret.
Post-checkReopen result, second scan, output usability, and unresolved items.
BoundaryBuild, environment, known limitation, and a clear human-review reminder.
This is an evidence record for a tested path, not a safety certificate or an advanced reporting feature.
Available now
See why visual covering is not removal.
Use the synthetic two-PDF demo to compare a black overlay with a comparison file authored without the canary values.