Simple and stable Node.js implementation of read-only S3 file system

Usage no npm install needed!

<script type="module">
  import s3fuse from 'https://cdn.skypack.dev/s3fuse';


s3fuse 📁☁️

s3fuse is yet another S3 FUSE file system implementation, this time in Node.js! It aims to be a super simple, stable and relatively fast way of browsing S3 and downloading files. Writing can be risky because of eventual consistency, and will not be supported by this project.

I wrote this quickly because I needed a really simple and stable S3 file system for running on a server application. Other options were either too complex, buggy or crashing/hanging all the time.


  1. Install Fuse
  2. Install Node.js
  3. Run npm install -g s3fuse


Specify AWS settings and credentials in your ~/.aws/ or in the environment:

export AWS_PROFILE=profile
export AWS_ACCESS_KEY_ID=key
export AWS_SECRET_ACCESS_KEY=secretkey

Then run:

s3fuse bucket-name /mnt/path


  • Caches dir and file names for faster browsing
  • Simple implementation and should be very stable


  • Read only
  • Only sequential requests (no parallelization yet in node-fuse-bidings)
  • Downloading files is very slow because of previous point


  • Handle objects with no permissions (currently will probably just give Input/output error)
  • Automated tests
  • Maybe use libfuse directly using node-ffi
  • Handle IsTruncated