wingetCollections
  • Auto
  • Search
  • Awesome Collections
  • Awesome Programs
  • My Favorite Collections
  • My Collections
  • Account / Settings
  • Terms of Service
  • Privacy
  • About
  • What is WinGet?

© 2026 winget.ragerworks.com

Loading...
Install ExoSnap using Winget - wingetCollections
  1. Go back
  2. Packages
  3. ExoSnap
ExoSnap logo

ExoSnap Codexo

av1 diagnostics nvenc overlay recorder screen-capture screen-recorder webcam
Use this command to install ExoSnap:
winget install --id=Codexo.ExoSnap -e

ExoSnap is a high-performance screen recorder designed for capturing applications, regions, or entire monitors with precision and reliability. Built around NVIDIA’s NVENC technology, ExoSnap delivers smooth 60 fps captures in formats like MKV, WebM, and MP4, while ensuring compatibility with modern codecs such as AV1 and HEVC.

Key Features:

  • Diagnostics-First Approach: ExoSnap includes a pre-flight readiness gate to identify potential issues before recording starts. It also provides live monitoring of frame drops, A/V drift, and disk usage, along with a post-recording report card for pipeline health analysis.

  • Multi-Track Audio Routing: Capture multiple audio sources simultaneously, including application audio, system sounds, and microphone input, with per-track gain control and noise suppression.

  • Webcam Overlay: Add live webcam video to your recordings with a mirrored preview for self-monitoring.

  • HDR Support: Detect HDR displays automatically and capture content as tone-mapped SDR (default) or native HDR10 for compatible setups.

  • NVIDIA NVENC Acceleration: Leverage hardware encoding for efficient, high-quality captures without taxing system resources.

Audience & Benefit:

Ideal for content creators, educators, and technical professionals who require reliable, professional-grade screen recordings. ExoSnap’s focus on diagnostics ensures that users can troubleshoot issues before they impact their work, while its flexible audio and video settings cater to a wide range of use cases. The software is particularly beneficial for those needing precise control over capture quality and performance.

ExoSnap can be installed via winget, making it easy to integrate into your workflow.

README

ExoSnap

Latest Downloads CI License

A diagnostics-first, Windows-native screen/application/region recorder. MKV-first capture with on-stop MP4 remux, a native NVENC pipeline, multi-track audio routing, HDR10 monitor capture, and a pre-flight readiness gate that blocks recording before a misconfiguration ruins it — no account, no analytics telemetry, and recordings are never uploaded or processed in the cloud (update checks against public GitHub Releases and consent-gated crash reports are the only opt-in network features; see PRIVACY.md).

Releases · Roadmap · Known limitations · Privacy

ExoSnap Record page

> Pre-1.0. ExoSnap is a Windows preview, not a finished 1.0. Settings, preset, and > recording-history file formats may change incompatibly between releases. > > NVIDIA NVENC is required. ExoSnap only supports NVIDIA hardware encoding today > (RTX 20-series or newer recommended). AMD, Intel, and software encoding are not available yet — > see the roadmap for the vendor-expansion plan. > > Windows 10/11 x64 only. Windows 11 is the primary target; Windows 10 is best-effort. > Distributed as a portable ZIP and an MSI installer, neither is code-signed yet, so Windows > SmartScreen may warn on first launch (see Code signing).

Versions
0.8.1
Website
github.com
License
Last updated
7/2/2026
Download latest version

Features

Reliability & recovery

  • Crash recovery — a recovery manifest is written before each session; a startup overlay offers Finish / Continue / Delete for each interrupted recording on the next launch. For MKV/WebM split recordings, segments that were already finalized before an interruption remain usable; an interrupted active segment may not be — see KNOWN_LIMITATIONS.md.
  • MKV-first recording with on-stop MP4 remux (progressive, faststart) via libavformat stream-copy; MKV and WebM are also available as direct output formats.
  • Low-disk guard — warns at a configurable soft threshold and hard-stops at a lower one, accounting for the transient MKV + output MP4 coexisting during remux.
  • FAT32 awareness — flags the 4 GiB per-file limit as a Diagnostics notice without blocking short recordings.
  • Container/codec compatibility registry — invalid combinations are rejected before recording starts, never mid-session.
  • Recording split (by time and/or size, first limit reached wins) with per-segment background MP4 remux.

Diagnostics as a feature

  • Pre-flight readiness gate — blockers and notices surface before you hit record, so a misconfigured capture is caught up front instead of failing mid-recording.
  • Typed FixAction model (Auto / Assisted / External) — every detected issue carries its safest concrete remedy, not just a warning string.
  • Live pipeline monitoring — frame drops, A/V drift, disk-fill ETA, and an on-screen size/drift overlay, with encoder-vs-capture-vs-disk-bound classification and root-cause correlation (e.g. VRR-vs-CFR judder).
  • Post-flight report card — frame-drop percentage, peak A/V drift, and pipeline health after every recording.
  • Opt-in PresentMon provider — elevation-gated tearing/game-present observation feeding the judder correlation, never a hard dependency.
  • Recording-error dialog — if a recording fails, a modal report explains why and, in official builds, offers the same consent-gated privacy-scrubbed crash-report channel.

Video & HDR

  • Native NVENC pipeline: H.264, HEVC, and AV1, with 10-bit (P010) output for HEVC Main10 and AV1.
  • Canonical rate-control model — Constant quality, Variable bitrate, Constant bitrate, mapped per encoder underneath; never mislabeled as x264/x265-style "CRF".
  • BT.709 color metadata on every output, plus a selectable Y'CbCr color range (Full/Limited) with a Diagnostics compatibility notice for players that ignore the range flag.
  • HDR displays are detected automatically. By default an HDR desktop records as tone-mapped SDR (BT.709) for universal playability. An expert "HDR handling" setting switches to native HDR10 recording — PQ/BT.2020, P010 10-bit, limited range, with mastering-display metadata carried into remuxed MP4. Native HDR10 requires HEVC or AV1 and covers both monitor and window/game capture (the latter via a scRGB FP16 frame pool); see KNOWN_LIMITATIONS.md for the exact boundary.
  • Webcam PiP overlay with live mirrored preview, DXGI-composited.

Audio

  • Multi-track routing with the default source order APP, SYS, MIC, each recorded to its own default track; any source can merge onto the track above it.
  • Per-track gain, mute, and a brickwall limiter (on by default) on the mixed bus.
  • Microphone DSP chain — high-pass filter, noise gate, AGC, and RNNoise neural noise suppression — each stage off by default and individually toggled.
  • Lossless PCM and FLAC (MKV-only) alongside Opus and AAC; configurable sample rate, channel count, and bit depth.

Presence & workflow

  • Capture-excluded on-screen overlays (recording status, diagnostics, countdown) via WDA_EXCLUDEFROMCAPTURE — visible to you, invisible to the recording.
  • Tray icon with idle/recording/paused states, an unread notification badge, and a notification hub alongside toast notifications.
  • Global hotkeys, single-frame capture, and an opt-in interactive quick-control pill overlay.
  • TOML recording presets with human-readable export/import and a manage dialog (rename, duplicate, delete, set default).
  • Four curated themes (two dark, two light) with instant single-click preview and selection; dark mode is the default.
  • Local-first crash capture (out-of-process Crashpad) with a consent-gated, privacy-scrubbed next-launch crash dialog.
  • In-app updates (Stable/Preview channels): a signature- and hash-verified download and in-place install via a dedicated updater process, with rollback on failure. Always off by default for self-built binaries, and every install step is visible to the user — never a silent auto-install.

See KNOWN_LIMITATIONS.md for the precise current support boundary and docs/roadmap.md for where this is headed.

Product defaults

SettingDefault
ThemeDark mode
ContainerMKV
Video codecAV1 (NVENC)
Audio codecOpus
Frame rateCFR 60 fps
Audio source orderAPP, SYS, MIC — context-aware defaults (APP only exists for a captured application window, enabled there; screen capture defaults SYS on, MIC off), each enabled source its own track
Navigation pagesRecord, Device, Settings, Diagnostics, Logs, About

Install / run

See README-PORTABLE.md for the portable-release quick start.

Runtime prerequisite: the Microsoft Visual C++ 2022 x64 Redistributable is required. It is normally already present on up-to-date Windows systems. If the app fails to start with a missing-DLL error, install it from .

The portable build is not code-signed. Windows SmartScreen may warn on first launch — this is expected (see Code signing).

Building from source

Prerequisites: Visual Studio 2022 (Desktop development with C++ workload), CMake 3.27+, Git.

Fast local development loop:

cmake --preset windows-x64-debug
cmake --build --preset windows-x64-debug-exosnap
scripts\run-tests.ps1 -Filter .
scripts\check-format.ps1

Full gate before merge:

scripts\check-format.ps1
git diff --check
cmake --build --preset windows-x64-debug
scripts\run-tests.ps1
scripts\check-quality.ps1 -StaticOnly
cmake --build --preset windows-x64-release-exosnap

scripts\run-tests.ps1 sets up the environment every test binary needs (throwaway EXOSNAP_CONFIG_DIR, offscreen Qt platform, Qt on PATH) and prints a compact summary plus any failing gtest cases; each CTest entry is one test binary, and -Filter/-R match binary names (e.g. recorder_core.), not individual gtest cases. Optional faster/cached builds are available via Ninja and sccache (winget install Ninja-build.Ninja / Mozilla.sccache). C++ code is formatted with clang-format and checked with clang-tidy; run scripts\pre-commit.ps1 before committing. See AGENTS.md for the full build-tooling reference.

Repo layout

app/          application source (UI, models, recording pipeline)
libs/         internal libraries (recorder_core, capability, etc.)
docs/
  roadmap.md  version roadmap and architecture guardrails
  decisions/  architecture decision records (ADRs)
packaging/    WinGet, Chocolatey, and Scoop packaging manifests
scripts/      build, quality, and release scripts
tests/        test targets
cmake/        CMake helper modules (VendorFFmpeg, etc.)
third_party/  vendored third-party sources (libmatroska, libebml, etc.)
tools/        developer tooling

Third-party / licensing

  • ExoSnap is licensed under GPL-3.0-or-later; see LICENSE.
  • ExoSnap bundles FFmpeg as LGPL-2.1-or-later shared DLLs (dynamic linking). Third-party component licenses are listed in THIRD_PARTY_NOTICES.md.
  • The FFmpeg binaries are produced by the companion repository Exoridus/exosnap-ffmpeg-build.

Code signing

ExoSnap participates in the SignPath Foundation free code-signing program for open-source projects, with code-signing infrastructure provided by SignPath.io. Once the certificate is issued, Windows release binaries (portable ZIP and MSI) will be signed; pre-certificate builds are unsigned, so Windows SmartScreen may warn on first launch.

Contributing / CI

GitHub Actions runs configure, build, and test on Windows for both windows-x64-debug and windows-x64-release, plus a lint pass. PRs are the review gate.

Work happens on branches cut from origin/main; main itself only ever moves by merging a PR. Two guards keep that true, both applied automatically by a CMake configure (or by running scripts/install-hooks.ps1 after a bare clone):

  • The pre-commit hook refuses a commit made while main is checked out. Committing on a local main is what makes it diverge — once those commits are squash-merged upstream the squash carries a new commit id, so the originals linger locally and every branch cut from that point drags them into its PR. Prefix a command with ALLOW_MAIN_COMMIT=1 for the rare legitimate case, such as a release bump.
  • pull.ff is set to only, so a git pull that would need a merge fails instead of quietly recording one. This is per-clone local config and cannot be committed, which is why the configure step sets it.

Branch from the remote ref rather than the local branch — git switch -c origin/main after a git fetch — so a stale local main cannot become the base of new work.

Links

  • Repository:
  • Issues:
  • Privacy: PRIVACY.md
  • Security: SECURITY.md