CloakScan v1.6.0 fixes four focused coverage and keyboard-accessibility problems.
What changed
- Enable all built-in rules now enables all 49 built-in detectors without replacing the active profile or dropping Cloak Lists, custom rules, formats, or session terms. CloakScan still clears the old scan result and requires a rescan.
- The disabled-rule explanation can collapse into a persistent coverage notice. It stays collapsed for the current session, reopens when the disabled rule set changes, and disappears when all relevant rules are enabled.
- Dismissing the general review reminder now lasts across navigation, scans, profile changes, output-mode changes, Back, and New Scan for the current session. Clear session or a reload restores it; the permanent review footer remains.
- The custom-term dialog now traps forward and reverse keyboard focus, closes with Escape from any control, restores focus to its opener, and makes background actions unavailable while open.
Downloads
- Windows:
CloakScan-Setup-1.6.0-x64.exe - Linux:
CloakScan_1.6.0_amd64.deborCloakScan_1.6.0_amd64.AppImage - Browser: live demo
The Windows installer is not Authenticode-signed, so SmartScreen may warn. The Windows and AppImage updater packages have detached signatures, and checksums are included for the downloads.
Verification
CI passed on 6d27977e24ca34b145ed99b4e5e989186c0242f8: 3,721 unit tests, 105 production browser tests, desktop checks on Windows and Linux, nine Rust tests, Rust formatting and lints, and zero npm-audit findings. Ten opt-in screenshot capture tests were skipped.
The Pages deployment, Windows package, and Linux packages were produced from the same commit. The Linux workflow inspected both packages and kept the AppImage alive for 20 seconds under Xvfb.
The deployed browser app reports v1.6.0, and the seven focused v1.6 production-flow regressions passed against the live Pages build. A separately identified, isolated Windows package installed as v1.6.0 and passed all four changed flows without touching the normal CloakScan installation or data.
Downloaded artifact checksums matched the workflow outputs. Both updater signatures verified against the public key embedded in CloakScan, and deliberately altered bytes were rejected.