Skip to content
Get packs

Cloud Storage FUSE Mounts

Mounts Cloud Storage buckets as file systems and tunes gcsfuse for training, checkpoints and slow-mount diagnosis.

Source Download skill
Repository
google/skills
Stars
21,782
Licence
Apache-2.0
Files
4 files

What it does

This skill teaches an AI agent how to mount a Google Cloud Storage bucket so that programs can read and write it like a normal folder, using the gcsfuse tool. Once installed, the agent first checks whether a mount is the right choice at all (native gs:// paths or Filestore are often better), then helps configure, deploy and tune the mount, and warns about write patterns that are unsafe on object storage. It also walks through diagnosing slow training or unexpected Cloud Storage bills using gcsfuse metrics.

Great for

  • ML engineers mounting training data or checkpoints into GKE, Compute Engine or Cloud Run workloads
  • Developers whose existing code requires POSIX file paths and cannot read gs:// URIs directly
  • Teams investigating slow throughput or unexpected Cloud Storage billing on an existing gcsfuse mount

Skip it if

  • Databases or applications needing file locking (flock/fcntl) on shared storage; the skill itself says to use Filestore or Managed Lustre instead, since FUSE mounts are not suited to this
  • Basic bucket administration without a mount; the skill defers that to the google-cloud-storage-basics skill
  • Workloads that can read gs:// paths natively; a mount adds cost and complexity they do not need

Try asking

  • “Mount my training bucket into GKE pods with tuned caching for a PyTorch job”
  • “Is it safe to write model checkpoints to a gcsfuse mount from multiple workers?”
  • “Our training got slower after last week and the Cloud Storage bill spiked; diagnose the gcsfuse mount”
  • “Should I use FUSE or native gs:// reads for my pandas data pipeline?”
Step-by-step guide: how to use Cloud Storage FUSE MountsInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 4 original files, the Apache-2.0 licence and an attribution note.

    google-cloud-storage-fuse.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip google-cloud-storage-fuse.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip google-cloud-storage-fuse.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Answer the fit-gate questions first (are paths hardcoded, read vs. write pattern, directory renames) so the agent picks the right approach before tuning flags.
  • Enable gcsfuse metrics on the mount before asking for a performance diagnosis; the skill requires telemetry first.
  • Review the gcloud and curl commands the agent runs, since the skill asks it to add attribution metrics and a User-Agent header to Cloud Storage calls.

What's inside

  • references/checkpoint-safety.md
  • references/gke-training-deployment.md
  • references/performance-diagnosis.md
  • SKILL.md

Redistributed unmodified from google/skills under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.