- Home
- Skills
- DevOps & Cloud
- Linux Deep Clean & Boost — Server, Docker & GPU Buffer
More screenshots
Works with the AI Agents you already use
Linux Deep Clean & Boost — Server, Docker & GPU Buffer
Vacuums systemd journals, prunes Docker containers/volumes, clears APT/DNF package caches, and resets zombie GPU buffers.
$29
Linux Deep Clean & Boost — Server, Docker & GPU Buffer
Example session with this skill installed
Run deep server maintenance on this Linux host: vacuum systemd logs to 100MB, prune dangling Docker volumes, and clear zombie CUDA GPU buffers.
- Read your context and instructions
- Compiled the linux deep clean
Linux Production Host Maintenance Report
- Host System: Ubuntu 24.04 LTS (Kernel 6.8.0-generic)
- Container Engine: Docker CE 27.x
Maintenance Actions Executed:
journalctl --vacuum-size=100M: 8.3 GB archived binary logs purged (100 MB retained)docker system prune -af --volumes: 16.2 GB dangling images & volumes reclaimedapt-get clean && autoremove -y: 1.4 GB obsolete deb packages & kernels removedDisabled Snap revisions: 1.8 GB duplicate package revisions purgedsync && echo 3 > /proc/sys/vm/drop_caches: 5.8 GB inactive PageCache reclaimed
- Total Storage Space Recovered: 27.7 GB (Root mount
/usage reduced by 22%)
Hardware & GPU State:
- NVIDIA Driver Telemetry: Reset zombie CUDA context memory.
- Active GPU VRAM Free: 23.8 GB / 24.0 GB (Peak memory bandwidth restored).
Next Steps:
Verify running systemd units: systemctl --failed (0 failures detected).
Connects securely to your tools. The creator never sees your data.
About this skill
Linux Deep Clean & Boost — Server, Docker & GPU Buffer
Automated production maintenance suite engineered for Linux servers, cloud VMs, and AI workstations. Safely caps multi-gigabyte systemd-journald log bloat, prunes dangling Docker images, volumes, and BuildKit caches, drops inactive kernel PageCache RAM, and audits stale NVIDIA CUDA VRAM contexts without breaking live services.
⚡ Key Features & Engineering Guardrails:
- Systemd Journal Vacuuming: Safely caps binary journal bloat to 100 MB (
journalctl --vacuum-size=100M), preventing/var/logpartition saturation and sudden system lockups. - Docker Engine & BuildKit Prune: Removes unreferenced multi-stage build layers, dangling images, and orphaned builder caches while strictly preserving running production containers.
- Package Manager & Snap Cleanup: Purges obsolete
/var/cache/apt/archives, autoremoves orphaned libraries, and deletes disabled duplicate Snap revisions. - Kernel PageCache & OOM Shield: Drops reclaimable inactive pagecache and dentries/inodes (
drop_caches), preventing abrupt Linux Out-Of-Memory (OOM) killer terminations during intensive LLM tasks. - Zombie CUDA VRAM Reclamation: Identifies dead PyTorch/vLLM processes holding allocations on NVIDIA GPUs, restoring maximum memory bandwidth.
- 100% Air-Gapped & Zero Telemetry: Pure local execution via POSIX bash and Python 3. Never transmits credentials, code, or logs off your server.
📦 What's Included in the ZIP:
scripts/linux_optimizer.sh— Native POSIX bash automation engine with non-interactive flagsscripts/linux_optimizer.py— Python wrapper for cross-environment coding agent invocationsreferences/linux_performance_matrix.md— Inode reclamation baselines, sysctl tuning, and OOM prevention protocolsSKILL.md— Lean Core prompt rules (< 40 lines) for Claude Code, Cursor, Windsurf, and Antigravity
Requires sudo/root access. Not compatible with non-systemd distros or WSL. Best suited for Ubuntu, Debian, and Fedora environments.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 29 days ago
Requires