Your browser is no longer supported

MangoDisk needs a newer browser to display this website correctly. Upgrade Microsoft Edge or open this page in another modern browser.

You can still download MangoDisk directly:

Upgrade Microsoft Edge

Deep Cleanup

Clean caches and build artifacts with MangoDisk, choose extra disks, or create custom cleanup rules by folder, file name, size, and modification time

Updated 2026/09/08

Deep Cleanup brings cleanable content from different locations into one review screen. You can see how much space each category uses before choosing what to remove.

What it can find

The exact results depend on your operating system, installed apps, and how the computer is used. Common categories include:

  • system, user, and application caches;
  • temporary website data created by browsers;
  • files left behind by removed applications;
  • AI models and transfer caches;
  • developer tool caches, Xcode data, and project build artifacts;
  • application optimization items and unused container build caches.

These items are not all selected automatically. MangoDisk groups and labels them by purpose and risk so you can review them efficiently.

Choose a scan type

Open the arrow next to the scan button to choose:

  • Standard Scan: checks common cleanup items and development projects for the current user. A good starting point for routine cleanup.
  • Choose Disks: the system disk is always scanned. Select additional disks to find development projects and build artifacts there, without searching every disk.
  • Custom Cleanup: finds files in selected folders using your matching conditions, useful for old logs or temporary files in known locations.

Set up custom cleanup rules

Open Custom Cleanup, add a named rule, and select or drag in its scan folders. Each rule supports:

  • File name patterns: for example, *.log, *.tmp. Use * for any number of characters, ? for one character, and commas between patterns.
  • Size and modification time: set minimum and maximum sizes, or find files last modified before or within a specified number of days.
  • Include subfolders: turn this off to check only files directly inside the selected folders.
  • Remove empty folders: remove empty subfolders after cleanup while keeping the selected folders themselves.

For example, search a specific app log folder for *.log files not modified in over 30 days, then review the results before cleaning. A name match does not mean a file is unnecessary; avoid broad rules in folders containing personal data.

Save rules for reuse or choose Save and Scan to search immediately. Enable Also run a standard scan to check common cleanup items at the same time. Scanning does not delete files: you still select the results and confirm cleanup.

Choose what to clean

After the scan, you can use these selection modes:

  • Smart recommendation prioritizes lower-risk content that can usually be recreated.
  • Select all selects every cleanable item in the current result.
  • Select none clears the current selection.
  • Manual selection appears after you adjust individual items.

“Low risk” usually means the content can be recreated. “Needs review” marks content worth opening and checking. These labels help with review; they do not make the decision for you.

Confirm cleanup

The footer shows selected items, estimated space to free, and any apps that need to be closed. Save your work and close the listed apps first. Items still in use are safely skipped.

Select Clean to review the chosen items and their impact one more time. After cleanup, the result is available in Cleanup History.

Not sure? Get an AI explanation

Not sure whether clearing a cache means downloading or generating it again? Hover over the item and click AI explanation to understand what happens after cleanup.

You decide whether to proceed; AI never takes action for you. Use the free service without an API key or connect your own model. See the AI explanations guide to get started.

Explanations use information about the selected item, but AI cannot decide which files matter to you. Custom rules in Deep Cleanup do not currently support AI explanations.