Piggydrive: Access Cloud Storage on Linux When OAuth Blocks You

If you’ve ever tried to access enterprise or university cloud storage from Linux, you know the pain. Institutional authentication policies often block third-party clients — rclone, native CLI tools, even WebDAV mounts all hit the same wall: authentication rejected. The organization doesn’t trust them.

I needed cloud storage access from Linux for research projects, course materials, and collaboration work. Every standard CLI tool I tried failed at authentication. So I built Piggydrive: a workaround that doesn’t fight the restrictions — it goes around them.

---

The Problem: Authentication is a Wall

Enterprise and university cloud providers design their authentication to block anything they don’t explicitly approve. For many organizations, that means:

• rclone: Fails with “application not permitted” errors
• Native CLI tools: Authentication token exchange rejected
• WebDAV mounts: Blocked at the policy level
• Direct API calls: Same authentication blocker

The issue isn’t your code. It’s the organization’s policy. You can’t fix it with better authentication logic.

---

The Solution: A Trusted Bridge

Piggydrive’s core idea is simple: use a machine that’s already trusted.

Many of us have a personal laptop (Mac or Windows) that runs the official cloud storage client — the organization trusts it completely. That machine syncs the entire cloud folder locally. All we need is a way to access those files from Linux without re-authenticating.

This is especially useful in university and corporate environments. Institutions often provide you with a machine (laptop or desktop) that’s already configured to access their cloud infrastructure. That machine is trusted by default. Instead of fighting to authenticate from Linux, use that trusted machine as a bridge.

The architecture:

Linux machine          Trusted laptop (Mac/Windows)
     │                      │
     │  Private network     │
     ├──────────────────────┤
     │                      │
piggydrive CLI    ←  Cloud client (official)
     │                      │
     │                      └─► Cloud folder (synced)
     │                          (all files available locally)

The laptop runs a small daemon that exposes filesystem operations over a secure channel. The Linux piggydrive CLI sends commands, the laptop daemon executes them on the synced cloud folder, and returns results. No authentication dance on Linux — the laptop already authenticated.

---

How It Works (Technical Overview)

The Bridge

The laptop daemon is a Python server that:

1. Listens on a private network — secure connection between trusted devices
2. Executes filesystem operations — list, read, write, move, delete
3. Handles cloud sync — triggers download of cloud-only files when needed
4. Returns structured data — clean API for the CLI to parse

The key insight: I’m not building a cloud storage client. I’m building a remote filesystem proxy. The laptop’s official cloud client does all the heavy lifting — authentication, sync, conflict resolution. The daemon just exposes the local folder.

The CLI

The Linux piggydrive CLI provides a clean interface:

# Search for files (fast, even on large folders)
piggydrive find "project" --max 50

# List a directory
piggydrive ls /Projects/research

# Pull a file to local disk
piggydrive pull "/Cloud/research/paper.pdf" ~/work/paper.pdf

# Push a file to cloud storage
piggydrive push ~/work/report.md "/Cloud/reports/report.md"

# Check sync status
piggydrive sync-status

Fast Search

The killer feature: piggydrive find uses the laptop’s built-in search (Spotlight on Mac, Windows Search on PC) for instant results. Even on a 200,000+ file tree, results come back in under a second. It searches filenames, not contents, and works on cloud-only placeholders — no need to download files just to find them.

---

Network Flexibility

We built Piggydrive using Tailscale for the private network connection — it’s simple to set up and works great across different networks. But the key requirement is just that the two devices can maintain a direct connection. Any VPN, mesh network, or direct SSH tunnel will work. Tailscale is just the tool we chose for convenience.

---

Real-World Usage Patterns

Pattern 1: Find and Download Research Files

# 1. Find all files related to a project
piggydrive find "research" --max 100

# 2. Filter to the file types you need (e.g., PDFs)
# 3. Download each one to your local workspace
piggydrive pull "/Cloud/research/document.pdf" ~/work/document.pdf

Pattern 2: Save Work to Cloud

# Generate a report locally
python generate_report.py > ~/work/report.md

# Push to cloud for access from other devices
piggydrive push ~/work/report.md "/Cloud/reports/report.md"

# Verify sync completed
piggydrive sync-status

Pattern 3: Quick Read Without Download

# Read a small file directly (downloads to temp, prints to screen)
piggydrive cat "/Cloud/notes.md"

---

Why This Approach Works

No Authentication Fighting

The laptop’s official client handles all authentication. Linux never touches authentication tokens. The organization’s policies don’t matter — the laptop is already trusted.

Works with Cloud-Only Files

Modern cloud storage uses “files on demand” — files exist as placeholders until opened. Piggydrive handles this transparently. Search finds placeholders instantly. Download triggers the actual file retrieval automatically.

Fast Search

Built-in search on Mac and Windows is extremely fast. Searching a large file tree takes under a second. Compare that to recursive directory traversal (which would timeout) or remote listing (which requires downloading everything).

Automation-Friendly

The CLI outputs structured data, has stable error codes, and is designed for automation. It works well in scripts and with agent workflows.

---

Limitations (Being Honest)

• Requires a laptop bridge: The daemon runs on macOS or Windows. You need a trusted machine already syncing cloud storage.
• Private network dependency: The bridge and CLI communicate over a private network. Both machines must be connected.
• Bridge must be online: If the laptop is asleep or offline, the CLI can’t reach it.
• Not a full cloud client: You can’t change cloud settings, manage sync rules, or do admin tasks. It’s a filesystem proxy, not a replacement for the official client.

---

Call to Action

• Was it fully vibe-coded? Yes.
• Does it solve my problem? Absolutely.
• Would I use it again? I already do — daily.

Try it: The code is open source at github.com/ilyasst/piggydrive. If you’re blocked by authentication restrictions, Piggydrive might be your workaround.

Contribute: Found a bug? Want a feature? The daemon and CLI are straightforward Python — fix it with your favorite tools.