Darkelf Shadow — Community Edition 7.0.27
Privacy-focused browsing with Qt WebEngine, local MiniAI analysis and Smart Canvas protection.
Darkelf Shadow combines network filtering, fingerprint defenses and session-focused browsing. Available through Python/PyPI and as a native macOS ARM64 application.
What's new in 7.0.27
- Native-looking black macOS window: a frameless Qt main window replaces the gray system title bar with custom red, yellow and green traffic-light controls. The browser retains its existing toolbar and tabs.
- Window arrangements: click the green button to toggle fullscreen, or hover for approximately 1.8 seconds to open a dark rounded arrangement panel. Position the current Shadow window in the left or right half, any of three columns, or one of four screen quarters. Maximizing/restoring is also available. This does not automatically tile other applications.
- Floating panels: Settings and Darkelf Inspector now use separate black rounded panels instead of traditional dialog-style windows.
- Refined Settings: Appearance layouts fit the floating panel, and About uses the available width more effectively. The About version label displays 7.0.27.
- Accent-aware controls: the arrangement panel reflects the selected browser accent, with custom-painted traffic lights and a delayed hover menu to prevent accidental opening.
Earlier 7.x improvements retained
- Faster filter startup: cached list downloads and merged subscriptions, direct matching for simple rules, and regex compilation for complex rules. Local filter startup fell from approximately 38 seconds to 6 seconds with about 417,000 network rules; timings vary by system and cache state.
- CNN playback fix: unsupported cookie-modification rules are skipped instead of blocking requests. Scriptlet and other unsupported page-action rules no longer become network blockers. CNN10 playback was verified in developer testing.
- Independent Darkelf site boundaries: embedded rules replace the copied Public Suffix List, with no separate
.datfile or suffix download. Coverage is curated rather than worldwide; unknown namespaces use exact-host comparison and can cause extra blocking between related subdomains. Unlisted shared-hosting boundaries remain a coverage gap. - Browser usability: video fullscreen hides browser controls and restores them on exit; delayed keyboard-filter installation checks for deleted Qt views.
- Quieter operation: repeated canvas messages are reduced and automatic terminal threat reports are removed. Optional diagnostics expose website JavaScript warnings/errors and matched network blockers.
Installation
Python / PyPI
Requires Python 3.11 or newer.
pip install --upgrade darkelf-shadow
darkelf-shadow
Native macOS ARM64
Download the DMG and matching checksum from GitHub Releases. The native release workflow is designed to include Developer ID signing, hardened runtime, notarization and stapling; verify these steps for the published build.
| Distribution | Engine and session model |
|---|---|
| Python / PyPI | Platform PySide6 / Qt WebEngine; off-the-record profile. The custom macOS engine is not included. |
| macOS ARM64 DMG | Custom Darkelf Qt WebEngine 6.11.2; named profile for native authentication, memory HTTP cache and nonpersistent cookies. |
The DMG's named profile is not off the record. Native authentication configuration and credentials can persist separately from browsing-session cookies. Downloads, filter caches and other saved files can also remain on disk.
Privacy and filtering
- Indexed filtering using EasyList, EasyPrivacy and uBlock-derived sources, plus declarative tracker-host checks and scoped compatibility exceptions.
- URL tracking-parameter cleanup and compatibility-aware cosmetic filtering.
- Local MiniAI threat scoring, fingerprint-event monitoring and response modes, without a cloud AI dependency.
- Canvas, WebGL, audio and other fingerprint mitigations. The custom DMG engine includes native WebGL modifications and disabled WebRTC; engine-specific patches do not transfer through standard PyPI dependencies.
The filter engine supports a subset of upstream rule syntax. Unsupported actions are skipped rather than implemented as request blockers.
Smart Canvas
| Mode | Behavior |
|---|---|
| BLOCKED | Canvas readback is disabled. |
| PROTECTED | Readback uses Darkelf's domain-sensitive noise. |
| TRUSTED | Native readback is permitted for trusted compatibility cases. |
Supported human-verification flows can receive an automatic, temporary session trust grant to reduce verification loops. Closing Darkelf clears that grant.
Authentication and media
The custom macOS engine includes native WebAuthn integration and H.264/AVC support. Touch ID/passkey availability depends on the signed app's keychain access group, provisioning, platform and website compatibility. Apple-specific authentication issues are not claimed resolved by 7.0.27.
H.264 support does not supply DRM support. Protected streams may require a compatible DRM module; Widevine is not bundled with Darkelf.
Optional diagnostics
Normal launches keep routine diagnostics quiet. To investigate website failures:
DARKELF_DIAGNOSTICS=1 darkelf-shadow
For a source checkout:
DARKELF_DIAGNOSTICS=1 python3.11 main.py
This prints website JavaScript warnings/errors and blocked requests with their matching rules. Launch without the variable to return to quiet operation.
Verify the macOS download
Place the DMG and checksum in the same directory, then run:
shasum -a 256 -c Darkelf-Shadow-7.0.27.dmg.sha256
Expected result:
Darkelf-Shadow-7.0.27.dmg: OK
License and scope
Darkelf Shadow is licensed under LGPL-3.0-or-later. The native app packages notices at:
Darkelf Shadow.app/Contents/Resources/LICENSEDarkelf Shadow.app/Contents/Resources/THIRD_PARTY_NOTICES.txt
Bundled Qt, Chromium, FFmpeg and other components retain their respective licensing requirements.
Provided AS IS, without warranty. Darkelf does not guarantee anonymity, zero disk traces or protection against every threat, and does not replace operating-system security.
Author and acknowledgments
Dr. Kevin Moore · Darkelf Project — Shadow Edition · 2025–2026
Thanks to the Mecha Comet Team and everyone testing and supporting Darkelf Shadow.
Metadata
Release files for darkelf-shadow 7.0.27
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| darkelf_shadow-7.0.27.tar.gz | 315.7 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| darkelf_shadow-7.0.27-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 639.8 kB
Release files / darkelf_shadow-7.0.27.tar.gz
| Download URL | darkelf_shadow-7.0.27.tar.gz |
|---|---|
| Size | 315.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
b7b2a6690a0cdf7aca87a6f2cec640e71d9c4d248b9096773ca67b5b442bf0a5
|
|
BLAKE2b-256 checksum How to use checksums |
ddb3dd6d9bfe141d1b24552e3318089a01c7984b145f3aa4c8a831164f8b986e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 14, 2026.
Transparency logRelease files / darkelf_shadow-7.0.27-py3-none-any.whl
| Download URL | darkelf_shadow-7.0.27-py3-none-any.whl |
|---|---|
| Size | 324.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
44bd6405376e7f715320c6123892ca9126e341194ecff45b31d0c0be7923ec67
|
|
BLAKE2b-256 checksum How to use checksums |
80836653f04fcac3e3d5f3a855b72c14043f8fc0989c8e40141a3e84749bf22c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 14, 2026.
Transparency log