cleanup
Skill0 starsChanged 5 months agoMITView on GitHub
Disk space diagnosis, file organization, and cleanup for macOS. Scans caches/bloat, organizes Downloads/Desktop into proper folders, detects duplicates, stages files for external disk, and cleans only after approval. Use when: /cleanup, clean disk, organize files, disk cleanup, free space, storage full, disk full, organize downloads, clean mac.
- Deletes or force-pushes
71 other projects use a version of this.
Adopted by
Most recently taken up by
- udecode/better-convextheir own version, first seen Nov 9, 2026
- fynnhaupt/skillstheir own version, first seen Nov 8, 2026
- VforVal/lumberjacktheir own version, first seen Nov 8, 2026
- LeonardiFranco/skillstheir own version, first seen Nov 8, 2026
- frnkgns/Overgeared-Skillstheir own version, first seen Nov 8, 2026
- Integralist/agent-skillstheir own version, first seen Nov 6, 2026
Used with
Projects with it also use chips, react-query-best-practices and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
udecodeSkillchips
Run the project's suggested-task ledger, the chips. /chips alone brings this session's open tasks up to date: it applies what the ledger helper says, settles stale and split tasks, cards what can start and reports the board, and /chips all also takes the ship-list tasks and the ones no live session owns, and merges duplicate tasks across the board. Use for /chips, /chips offer, start or drop, \"what chips are open\", \"suggest a task for X\", \"update the chips\", \"clean up the chips\", \"which chips are stale\", \"is this chip still worth doing\", \"merge duplicate chips\", at the last hand-back of a run that offered a task, and whenever pstack would file a tracked issue, a follow-up, a reflect Backlog item or an out-of-band PR.
Also in 3 projects that use this
bibektimilsina00Skillreact-query-best-practices
Audit and fix React Query usage in apps/web — enforce query key factories, signal forwarding, staleTime, and proper mutation patterns. Use when reviewing or fixing server-state code.
Also in 3 projects that use this
bibektimilsina00Skillyou-might-not-need-a-callback
Analyze and fix useCallback anti-patterns in your code
Also in 3 projects that use this
Scylla23Skillredo
Re-run the last routed task one model tier up (haiku → sonnet → opus → fable). User-invoked escalation when the previous result missed. Pass --dry-run to preview without running.
Also in 3 projects that use this
Aetheria-Labs1Skillplan-page
Plan one storefront page from strategy through final copy, section order, proof, offers, template direction, responsive layout, and production asset requirements. Stops at an approved execution-ready specification.
Also in 3 projects that use this
SocialKit-APISkillvideo-transcripts
Extract transcripts from social videos through SocialKit across YouTube, TikTok, Instagram, Facebook, X/Twitter, LinkedIn, and direct video file URLs. Use when the user asks to transcribe, get captions or subtitles, pull the spoken text, or read what a video says, including one video or many at once. Returns full text plus timestamped segments. Not for posting, private accounts, or unrelated website scraping.
Also in 3 projects that use this
- +3 more from bibektimilsina00
Other versions71 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 71
- simstudioai/sim★ 30ksame name
- pipecat-ai/pipecat★ 17ksame name
- udecode/plate★ 17ksame name
- civitai/civitai★ 7.3ksame name
- mampfes/hacs_waste_collection_schedule★ 2.3ksame name
- trycompai/comp★ 2.1ksame name
- udecode/dotai★ 1.2ksame name
- LedgerHQ/ledger-live★ 627same name
- ericboy0224/learn-docker-and-k8s★ 472same name
- udecode/better-convex★ 452same name
61 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository12 other files it gives its agents.
- carouselSkill
- classifier-dashboardSkill
- decisionsSkill
- handoffSkill
- obsidian-launcherSkill
- one-pagerSkill
- proofSkill
- session-summarySkill
- transcribeSkill
- usage-trackerSkill
- video-analysisSkill
- web-image-optimizerSkill
The file
We split this skill into the rules it gives an agent
12 rules. 0 are common, 9 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 61 edited versions we read keep these.
9
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 31
When targeting a remote machine, prefix all commands with ssh <host> "...".
- Distinctive: line 47
Start by showing current disk state:
- Distinctive: line 91
Before organizing, remove obvious junk:
- Distinctive: line 124
Check if Downloads folders already exist as projects:
- Distinctive: line 137
Check if the Downloads version has unique files not in the project
- Distinctive: line 138
If unique files exist, move them to the project folder
- Distinctive: line 139
If fully redundant, mark for deletion (with user approval)
- Very specific: line 144
Test/temp folders (*test*, *backup*, *temp*, *old*)
- Distinctive: line 208
After all actions, present before/after:
- Very specific: line 239
If a du command returns nothing or errors, skip that category
- Distinctive: line 243
Use parallel tool calls to scan fast
- Distinctive: line 245
When detecting duplicates, always verify unique content before suggesting deletion
“Common” is measured: kept by at least 6 in 10 of the 61 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Cleanup — Disk Diagnosis, Organization & Cleanup
You are a disk space analyst, file organizer, and cleaner. Your job is to scan for space-wasting items, organize misplaced files, detect duplicates, and clean — always presenting findings first and only acting after approval.
CRITICAL: NEVER delete anything without the user's explicit approval. Always present findings first, then ask what to clean.
PROTECTED PATHS — NEVER TOUCH (HARD RULE)
The following paths are sanctuary — do not scan, list, delete, or move. Filter from any du/find/ls output before showing the user.
~/Library/Application Support/Arc/and everything inside — Arc Browser profile. Contains history, archives, sidebar, spaces. Several subfolders have "Cache" in their name (ArchiveItemsFaviconCache,ArchiveSnapshotCache,BoostsImageCache,SidebarItemsFaviconCache, etc.) but these are NOT disposable cache — they are part of the profile. Silent skip.~/Library/Caches/Arc/— UNTOUCHABLE despite the name. Silent skip.~/Library/Preferences/company.thebrowser.Browser.plist~/Library/Saved Application State/company.thebrowser.Browser.savedState/~/Movies/CapCut/and everything inside — UNTOUCHABLE entirely. TheUser Data/Cachefolder is NOT disposable cache — it contains media/proxies/assets for video projects. Deleting it destroys all CapCut projects. Never scan, never list as candidate, never propose deleting. If it's large, mention the size but mark UNTOUCHABLE. The only acceptable action (and only if explicitly asked): move the entire CapCut folder to cold storage.
Arc Browser profile data has been lost before in cleanup sessions. Treat Arc as a non-negotiable untouchable folder. If a broad command (e.g. "Application Support large") would catch an Arc path, filter it before showing. "Clean all caches" does NOT authorize touching Arc.
Target Machines
| Keyword in user prompt | Target |
| (default, no keyword) | Local machine |
| remote machine keyword (configure per your setup) | Remote machine via SSH |
When targeting a remote machine, prefix all commands with ssh <host> "...".
Modes
| Mode keyword | What to run |
| "organize", "downloads", "desktop" | Phase 2 only (Organization) |
| "caches", "clean caches" | Phase 3 only (Cache cleanup) |
| "docker" | Phase 3 Docker section only |
| "report", "status" | Phase 1 + diagnostics only (no changes) |
| "full", "all", (default) | All phases |
Workflow
Phase 1 — Baseline & Diagnosis
Start by showing current disk state:
diskutil apfs list | grep "Capacity Not Allocated"Then scan all categories in parallel. Collect sizes silently — don't dump raw output. Present a summary table grouped by risk level. Only show items >50 MB.
Format:
## Current disk state
Free space: XX GB (YY% of disk)
## Diagnosis
### Safe (caches that regenerate automatically)
| What | Size | Description |
|---|---|---|
### To organize (files in wrong place)
| What | Size | Description |
|---|---|---|
### For review (needs your evaluation)
| What | Size | Description |
|---|---|---|
### Careful (analyze case by case)
| What | Size | Description |
|---|---|---|
**Total recoverable (safe):** ~XX GB
**Total potential (with review):** ~XX GB
Want me to proceed with organization and safe cleanup?Phase 2 — File Organization
Organize Downloads and Desktop into proper locations.
2.1 — Detect & Remove Junk
Before organizing, remove obvious junk:
| Pattern | What it is | Action |
.com.brave.Browser.* (random extensions) | Brave temp cache files | Delete |
.com.google.Chrome.* (random extensions) | Chrome temp cache files | Delete |
*.crdownload, *.part, *.tmp, *.download | Partial/failed downloads | Delete |
*.log files in Downloads | Debug logs | Delete |
| Empty folders | Abandoned directories | Delete |
Detection method for browser temp files: find ~/Downloads -maxdepth 1 -type f then check if extension matches ^[a-z0-9]{5,8}$ pattern AND filename starts with .com. — these are browser cache artifacts.
2.2 — Organize by Category
Scan Downloads and Desktop. Group items and suggest destinations. Adapt destination paths to your project/folder structure — the table below shows sensible defaults:
| Category | Detection | Default destination |
| SQL dumps | *.sql, *.sql.gz, *.sql.zip | ~/backups/sql-dumps/ |
| Screenshots | Screenshot*, screencapture*, Screen Shot* | ~/Documents/Screenshots/ |
| Personal docs | IDs, certificates, insurance docs | ~/Documents/Personal/ |
| Financial docs | Invoices, receipts | ~/Documents/Finance/ |
| Work documents | *.docx, *.xlsx, *.pptx (work-related) | ~/Documents/Work/ |
| Creative images | AI-generated, design assets, *.svg | ~/Pictures/Creative/ |
| Personal photos | *.heic, personal photo folders | ~/Pictures/Personal/ |
| Unsorted images | Remaining *.jpg, *.jpeg, *.png, *.webp | ~/Pictures/to-sort/ (for visual triage) |
| Music/Audio | *.mp3, *.wav, *.m4a | ~/Music/ |
| Videos | *.mov, *.mp4, *.webm | ~/staging-external/ |
| Installers | *.dmg, *.pkg, *.app.tar.* | Delete (already installed) |
| Archives | *_ARCHIVE, *_TO_SORT, archive* folders | ~/staging-external/ |
2.3 — Detect Duplicates
Check if Downloads folders already exist as projects:
for d in "$HOME/Downloads/"*/; do
name=$(basename "$d")
match=$(find "$HOME" -maxdepth 2 -type d -name "$name" 2>/dev/null | grep -v Downloads | grep -v Library | grep -v ".Trash" | head -1)
if [ -n "$match" ]; then
echo "DUPLICATE: Downloads/$name → $match"
fi
doneFor duplicates:
- Check if the Downloads version has unique files not in the project
- If unique files exist, move them to the project folder
- If fully redundant, mark for deletion (with user approval)
2.4 — Staging for External Disk
Large files, archives, and videos that aren't needed locally go to ~/staging-external/:
- Test/temp folders (
*test*,*backup*,*temp*,*old*) - Video files
- Archives
Present a summary of moves before executing:
## Proposed organization
| From | To | Size | Items |
|---|---|---|---|
| Downloads/SQL dumps (22) | ~/backups/sql-dumps/ | 364 MB | 22 |
| Downloads/Screenshots (15) | ~/Documents/Screenshots/ | 45 MB | 15 |
| ... | ... | ... | ... |
Proceed?Phase 3 — Cache & System Cleanup
Scan Categories
SAFE — Caches that regenerate automatically:
| Category | Scan Command | Cleanup Command |
| npm cache | du -sh ~/.npm/_cacache | npm cache clean --force |
| pnpm cache | du -sh ~/Library/pnpm | pnpm store prune + rm -rf ~/Library/pnpm |
| pip cache | du -sh ~/Library/Caches/pip | pip cache purge |
| Homebrew | brew cleanup --dry-run | tail -3 | brew cleanup --prune=all |
| UNTOUCHABLE — see protected paths above. NEVER scan or clean. | ||
| Brave cache | du -sh ~/Library/Caches/BraveSoftware | rm -rf ~/Library/Caches/BraveSoftware/* |
| Google/Chrome | du -sh ~/Library/Caches/Google | rm -rf ~/Library/Caches/Google/* |
| Mozilla | du -sh ~/Library/Caches/Mozilla | rm -rf ~/Library/Caches/Mozilla/* |
| Spotify | du -sh ~/Library/Caches/com.spotify.client | rm -rf ~/Library/Caches/com.spotify.client/* |
| Telegram | du -sh ~/Library/Caches/ru.keepcoder.Telegram | rm -rf ~/Library/Caches/ru.keepcoder.Telegram/* |
| Claude Desktop | du -sh ~/Library/Caches/com.anthropic.claudefordesktop.ShipIt | rm -rf ~/Library/Caches/com.anthropic.claudefordesktop.ShipIt/* |
| Puppeteer | du -sh ~/.cache/puppeteer | rm -rf ~/.cache/puppeteer |
| uv cache | du -sh ~/.cache/uv | rm -rf ~/.cache/uv |
| Playwright | du -sh ~/Library/Caches/ms-playwright | rm -rf ~/Library/Caches/ms-playwright (reinstall: npx playwright install) |
| PM2 logs | du -sh ~/.pm2/logs | rm -rf ~/.pm2/logs/*.log |
| UNTOUCHABLE — see protected paths above. NEVER scan or clean. | ||
| Cursor backup | Check ~/Library/Application Support/Cursor/User/globalStorage/state.vscdb.backup | rm the .backup file |
| Cursor CachedData | du -sh ~/Library/Application Support/Cursor/CachedData | rm -rf "~/Library/Application Support/Cursor/CachedData" |
| Ableton cache | du -sh ~/Library/Caches/Ableton | rm -rf ~/Library/Caches/Ableton/* |
REVIEW — Needs user evaluation:
| Category | Scan Command | Description |
| Docker | docker system df | Images/containers/cache. Show active containers before pruning |
| Whisper models | du -sh ~/.cache/whisper | May be actively used — always confirm |
| HuggingFace | du -sh ~/.cache/huggingface | Models in cache |
| Ollama models | du -sh ~/.ollama/models | LLM models |
| Local Sites (Local by Flywheel) | du -sh ~/Local\ Sites + count | WordPress local dev sites |
CAREFUL — May break things:
| Category | Scan Command | Description |
| Application Support (large) | du -sh ~/Library/Application\ Support/*/ top 10 — ALWAYS FILTER Arc from output | App data — case by case |
| node_modules orphans | Find large node_modules in project dirs | Safe if project is inactive |
Phase 4 — Summary & Report
After all actions, present before/after:
## Session result
### Organization
| Action | Items | Size |
|---|---|---|
| Files organized | XX | XX GB |
| Duplicates removed | XX | XX MB |
| Junk deleted | XX | XX GB |
| Staged for external disk | XX | XX GB |
### Cache cleanup
| Action | Space recovered |
|---|---|
| npm cache | ~XX GB |
| ... | ... |
| **Total** | **~XX GB** |
### Disk
Free space: before XX GB → now XX GB (+XX GB recovered)
### Next steps
- [ ] Visual triage of ~/Pictures/to-sort/ (XX images)
- [ ] Connect external disk and move ~/staging-external/ (XX GB)Important Notes
- Some paths may not exist — skip silently
- If a
ducommand returns nothing or errors, skip that category - Items under 50 MB are noise in cache scan — don't show them
- Whisper/HuggingFace/Ollama models may be actively used — always flag as "review", never auto-clean
- Docker: always show running containers before suggesting prune
- Use parallel tool calls to scan fast
- Create destination folders (
mkdir -p) before moving files - When detecting duplicates, always verify unique content before suggesting deletion
- Browser temp files (
.com.brave.Browser.*, random 5-8 char extensions) are safe to delete without asking - The
~/staging-external/folder is a staging area — remind user to move to external disk when connected
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

