Apps
Slack
Slack’s desktop app stores Chromium disk cache entries for channel images, emoji, downloaded UI assets, and API responses, along with Electron V8 code cache files that speed up JavaScript startup and GPUCache data containing compiled shader and rendering state for hardware-accelerated views. It also accumulates app logs that record launches, network activity, and client errors during daily use. Kudu removes Slack’s Cache_Data, Code Cache, GPUCache, and log files while leaving workspaces, messages, sign-in state, preferences, and downloaded files alone.
Why it matters
Why clean Slack?
- Corrupted Chromium disk cache entries can make channel avatars, custom emoji, or image previews load as blanks until Slack refetches them
- Stale cached web assets after a Slack update can leave the desktop client showing partial UI, missing panes, or a spinner that never finishes
- Invalidated Electron V8 bytecode after an app update causes a noticeably slow first launch while Slack rebuilds its Code Cache
- GPUCache data compiled against an older graphics driver can become invalid after a driver update, causing black windows, flicker, or choppy sidebar rendering
- An oversized Cache_Data folder keeps old workspace assets and API responses around, wasting disk space and making Slack revalidate large amounts of cached content
- Accumulated logs grow quietly over time and can make troubleshooting harder by burying current errors under old connection and startup noise
What gets cleaned
Cache paths Kudu targets
Windows
%AppData%/Slack/Cache/Cache_Data |
%AppData%/Slack/Code Cache |
%AppData%/Slack/GPUCache |
%AppData%/Slack/logs |
macOS
~/Library/Application Support/Slack/Cache/Cache_Data |
~/Library/Application Support/Slack/Code Cache |
~/Library/Application Support/Slack/GPUCache |
Linux
~/.config/Slack/Cache/Cache_Data |
~/.config/Slack/Code Cache |
~/.config/Slack/GPUCache |
Frequently asked
Common questions about Slack
Related cleaners
Free & open source
Download Kudu and reclaim your disk space.
Available on Windows, macOS, and Linux. No account required, no feature gates, no telemetry without consent. All cleaning targets are open source and community-auditable.