agentsclimarketplace

Disk cleanup

Skill ebal/AI-Skills/disk-cleanup

A portable, shareable collection of AI agent skills — structured instruction packs that teach coding agents how to perform specific tasks consistently.

Install
npx -y skills add ebal/AI-Skills --skill disk-cleanup

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

3 things to look at

  • 22 days oldThe repository was created 22 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Deploy and run a disk cleanup script on remote Debian/Ubuntu systems. Use when the user says "clean disk", "free space", "disk cleanup", "clean up server", "deploy cleanup", or wants to run /root/clean.sh on a remote host.

SKILL.md

3.0 KB, as published. Nobody here has run it

Disk Cleanup

Deploys and runs a disk cleanup script on remote Debian/Ubuntu systems. Removes old kernels, clears caches, vacuums logs, and prunes Docker/pip/npm — with before/after disk usage reporting.

Locating the script

Commands below use ${SKILL_DIR} for the folder that contains this skill. Set it to wherever the skill is installed, for example:

SKILL_DIR="$HOME/.claude/skills/disk-cleanup"     # Claude Code
# SKILL_DIR="$HOME/.config/opencode/skills/disk-cleanup"  # OpenCode / OpenWork
# On Claude Code you can also reuse the injected variable: SKILL_DIR="$CLAUDE_SKILL_DIR"

Quick Start

Deploy and run on a remote host:

# 1. Copy the script to the remote host
scp "${SKILL_DIR}/clean.sh" HOST:/tmp/clean.sh

# 2. SSH in and run as root
ssh HOST "sudo bash /tmp/clean.sh"

Or for a safe preview first:

ssh HOST "sudo bash /tmp/clean.sh --dry-run"

What It Does

StepWhat is cleaned
1Old kernel images (keeps the running kernel)
2Package documentation (/usr/share/doc/*)
3systemd journal (keeps last 7 days)
4apt cache (autoclean + clean)
5Unused dependencies (autoremove --purge)
6pip cache (if pip3 is installed)
7npm cache (if npm is installed)
8Docker unused images/containers/networks (if Docker is running)
9Filesystem sync

Prints disk usage before and after, with estimated MB freed.

Options

FlagEffect
--dry-runShow what would be deleted without touching anything
-h / --helpPrint usage info

Deploying to Multiple Hosts

To deploy to a list of servers:

HOSTS="web1 web2 db1"

for host in $HOSTS; do
    echo "=== $host ==="
    scp "${SKILL_DIR}/clean.sh" "$host:/tmp/clean.sh"
    ssh "$host" "sudo bash /tmp/clean.sh"
    echo ""
done

Or use a dry-run across all hosts first:

for host in $HOSTS; do
    echo "=== $host ==="
    scp "${SKILL_DIR}/clean.sh" "$host:/tmp/clean.sh"
    ssh "$host" "sudo bash /tmp/clean.sh --dry-run"
    echo ""
done

Scheduling via Cron

To run weekly (e.g., every Sunday at 3 AM):

# On the remote host:
echo "0 3 * * 0 root /tmp/clean.sh >> /var/log/disk-cleanup.log 2>&1" \
    > /etc/cron.d/disk-cleanup

Requirements

  • Debian/Ubuntu system with apt
  • Root access (script checks for $EUID -ne 0)
  • journalctl for log vacuuming (systemd-based systems only)

Notes

  • Journal vacuum keeps 7 days (configurable via JOURNAL_MIN variable at top of script)
  • Docker/pip/npm cleanup is skipped automatically if not installed
  • Safe to re-run — idempotent, no side effects on already-clean systems
  • The script syncs the filesystem at the end to flush all writes

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.