karlorz/cloudrouter/cloudrouter

Name: cloudrouter
Author: karlorz

Trust tier 3

community

Source check pending

Checking that the SKILL.md file is still in the repository.

Manage cloud development sandboxes with cloudrouter. Create, sync, and access remote VMs with GPU support, Docker, and browser automation. Use when asked to create a sandbox, spin up a dev environment, run code in the cloud, use GPUs, automate a browser, or interact with remote VMs.

Install

npx vskill@latest install karlorz/cloudrouter/cloudrouter

SKILL.md

cloudrouter - Cloud Sandboxes for Development

cloudrouter manages cloud sandboxes for development. Use these commands to create, manage, and access remote development environments with GPU support and browser automation.

When this skill is invoked

When the user invokes /cloudrouter or /cr without a specific task, present the available modes:

cloudrouter - Cloud Development Sandboxes
 
  Modes:
    cloudrouter start .                    Sync current directory to a cloud sandbox
    cloudrouter start --size small .       Smaller sandbox (2 vCPU, 8 GB)
    cloudrouter start --gpu T4 .           Sandbox with T4 GPU (16GB VRAM)
    cloudrouter start --gpu A100 .         Sandbox with A100 GPU (40GB VRAM)
    cloudrouter start --gpu H100 .         Sandbox with H100 GPU (80GB VRAM)
 
  Manage:
    cloudrouter ls                         List all sandboxes
    cloudrouter code <id>                  Open VS Code in browser
    cloudrouter jupyter <id>               Open Jupyter Lab in browser
    cloudrouter pty <id>                   Open terminal session
    cloudrouter vnc <id>                   Open VNC desktop
    cloudrouter stop <id>                  Pause sandbox
    cloudrouter resume <id>                Resume paused sandbox
 
  Browser automation:
    cloudrouter browser snapshot -i <id>       Get interactive elements
    cloudrouter browser open <id> <url>        Navigate to URL
    cloudrouter browser screenshot <id>        Take screenshot
 
  Run "cloudrouter start --help" for all options.

Setup

Before doing anything else, always run this command at the start of any cloudrouter task. Do not ask the user — just run it automatically:

bash

npm install -g @karlorz/cloudrouter   # Install/update CLI

This is idempotent — safe to run even if already up to date. This ensures the agent always has the latest CLI.

This installs both cloudrouter and cr (shorthand) as CLI commands.

Then authenticate:

bash

cloudrouter login

If the user hasn't logged in yet, prompt them to run cloudrouter login first before using any other commands.

Quick Start

bash

cloudrouter login                        # Authenticate (opens browser)
cloudrouter start .                      # Create sandbox from current directory
cloudrouter start --gpu T4 .             # Create sandbox with GPU
cloudrouter start --size small .         # Create smaller sandbox
cloudrouter code <id>                    # Open VS Code
cloudrouter jupyter <id>                 # Open Jupyter Lab
cloudrouter pty <id>                     # Open terminal session
cloudrouter ls                           # List all sandboxes

Preferred: Always use cloudrouter start . or cloudrouter start <local-path> to sync your local directory to a cloud sandbox. This is the recommended workflow over cloning from a git repo.

Commands

Authentication

bash

cloudrouter login               # Login (opens browser)
cloudrouter logout              # Logout and clear credentials
cloudrouter whoami              # Show current user and team

Creating Sandboxes

bash

# Standard sandbox (syncs local directory) — DO NOT pass --size, default is large (8 vCPU, 32 GB)
cloudrouter start .                        # Create from current directory (recommended)
cloudrouter start ./my-project             # Create from a specific local directory
cloudrouter start -o .                     # Create and open VS Code immediately
cloudrouter start -n my-sandbox .          # Create with a custom name
 
# Size presets — only use if the user specifically requests a different size
cloudrouter start --size small .           # 2 vCPU, 8 GB RAM, 20 GB disk — only if user asks
cloudrouter start --size medium .          # 4 vCPU, 16 GB RAM, 40 GB disk — only if user asks
cloudrouter start --size large .           # 8 vCPU, 32 GB RAM, 80 GB disk (DEFAULT — no flag needed)
cloudrouter start --size xlarge .          # 16 vCPU, 64 GB RAM, 160 GB disk — only if user asks
 
# With GPU (auto-selects Modal provider)
cloudrouter start --gpu T4 .               # T4 GPU (16GB VRAM)
cloudrouter start --gpu L4 .               # L4 GPU (24GB VRAM)
cloudrouter start --gpu A10G .             # A10G GPU (24GB VRAM)
cloudrouter start --gpu A100 .             # A100 GPU (40GB VRAM) - requires approval
cloudrouter start --gpu H100 .             # H100 GPU (80GB VRAM) - requires approval
cloudrouter start --gpu H100:2 .           # Multi-GPU: 2x H100
 
# With custom resources (override --size values)
cloudrouter start --cpu 12 --memory 49152 .  # Custom CPU and memory
cloudrouter start --disk 100 .               # Custom disk size (GB)
cloudrouter start --image ubuntu:22.04 .     # Custom container image
 
# From git repo (URL as positional arg or --git flag)
cloudrouter start https://github.com/user/repo    # Clone git repo directly
cloudrouter start --git user/repo                  # Clone via shorthand
cloudrouter start --git user/repo -b main          # Clone specific branch
 
# Provider selection
cloudrouter start -p e2b .                 # Use E2B provider (default)
cloudrouter start -p modal .               # Use Modal provider
 
# Custom timeout
cloudrouter start --timeout 1800 .         # 30-minute timeout (default: 600s = 10 min)

Size Presets

Size	vCPU	RAM	Disk	Notes
small	2	8 GB	20 GB	Light tasks
medium	4	16 GB	40 GB	Standard development
large	8	32 GB	80 GB	Default
xlarge	16	64 GB	160 GB	Heavy workloads

Do NOT pass `--size` unless the user explicitly asks for a specific size. The default is large (8 vCPU, 32 GB) which is appropriate for most tasks. Using --size small unnecessarily will make builds slower and may cause out-of-memory issues.

Individual resource flags (--cpu, --memory, --disk) override --size values.

GPU Options

GPU	VRAM	Best For	Availability
T4	16GB	Inference, fine-tuning small models	Self-serve
L4	24GB	Inference, image generation	Self-serve
A10G	24GB	Training medium models	Self-serve
L40S	48GB	Inference, video generation	Requires approval
A100	40GB	Training large models (7B-70B)	Requires approval
A100-80GB	80GB	Very large models	Requires approval
H100	80GB	Fast training, research	Requires approval
H200	141GB	Maximum memory capacity	Requires approval
B200	192GB	Latest gen, frontier models	Requires approval

GPUs requiring approval: contact founders@manaflow.ai.

Multi-GPU: append :N to the GPU type, e.g. --gpu H100:2 for 2x H100.

All `start` Flags

-n, --name <name>       Name for the sandbox
-o, --open              Open VS Code after creation
    --size <preset>     Machine size: small, medium, large (default), xlarge
    --gpu <type>        GPU type (T4, L4, A10G, L40S, A100, H100, H200, B200)
    --cpu <cores>       CPU cores (overrides --size)
    --memory <MiB>      Memory in MiB (overrides --size)
    --disk <GB>         Disk size in GB (overrides --size)
    --image <image>     Container image (e.g., ubuntu:22.04)
    --git <repo>        Git repository URL or user/repo shorthand
-b, --branch <branch>   Git branch to clone
-p, --provider <name>   Sandbox provider: e2b (default), modal
-T, --template <id>     Template ID — DO NOT use template names from `cloudrouter templates`; use --gpu flags instead
    --timeout <secs>    Sandbox timeout in seconds (default: 600 = 10 minutes)

Warning: Do NOT pass template names (e.g. cmux-devbox-base) to the -T flag. These are display names, not valid template IDs.

Aliases: cloudrouter start, cloudrouter create, cloudrouter new all do the same thing.

Managing Sandboxes

bash

cloudrouter ls                           # List all sandboxes
cloudrouter ls -p modal                  # List only GPU sandboxes
cloudrouter ls -p e2b                    # List only Docker sandboxes
cloudrouter status <id>                  # Show sandbox details and URLs
cloudrouter stop <id>                    # Pause sandbox (preserves state, can resume)
cloudrouter pause <id>                   # Same as stop — alias
cloudrouter resume <id>                  # Resume a paused sandbox
cloudrouter extend <id>                  # Extend sandbox timeout (default: +1 hour)
cloudrouter extend <id> --seconds 7200   # Extend by 2 hours
cloudrouter delete <id>                  # Delete sandbox permanently
cloudrouter templates                    # List available templates

Important: stop and pause are the same command — they preserve sandbox state. Use resume to bring a paused sandbox back. Use delete (aliases: rm, kill) to permanently destroy a sandbox.

Do NOT use `--timeout` with extend — the flag is --seconds.

Access Sandbox

bash

cloudrouter code <id>           # Open VS Code in browser
cloudrouter jupyter <id>        # Open Jupyter Lab in browser
cloudrouter vnc <id>            # Open VNC desktop in browser
cloudrouter pty <id>            # Interactive terminal session

Work with Sandbox

bash

cloudrouter pty <id>                       # Interactive terminal session (use for ongoing work)
cloudrouter ssh <id> <command>             # Execute a one-off command via SSH
cloudrouter ssh <id> "ls -la"              # Run a command (always quote the command string)
cloudrouter pty-list <id>                  # List active PTY sessions

Important: Prefer cloudrouter pty for interactive work. Use cloudrouter ssh for quick one-off commands.
CRITICAL: Always quote the command string. cloudrouter ssh <id> ls -la will FAIL because -la is parsed as a cloudrouter flag. Always wrap in quotes: cloudrouter ssh <id> "ls -la".

IMPORTANT: Fix npm Permissions Before Any npm Command

ALWAYS run this before `npm install` or any npm command in a new sandbox:

bash

cloudrouter ssh <id> "sudo chown -R 1000:1000 /home/user/.npm"

Without this, npm install will fail with EACCES/ENOENT errors on .npm/_cacache. This must be done once per sandbox, before the first npm operation. Do not skip this step.

File Transfer

Upload and download files or directories between local machine and sandbox.

Command signatures:

cloudrouter upload <id> [local-path] — accepts 1-2 positional args: sandbox ID and optional local path
cloudrouter download <id> [local-path] — accepts 1-2 positional args: sandbox ID and optional local path
Use -r <remote-path> flag to specify a non-default remote directory (default: /home/user/workspace)
Do NOT pass remote paths as positional arguments — this will error. Always use the -r flag.

bash

# Upload (local -> sandbox)
cloudrouter upload <id>                            # Upload current dir to /home/user/workspace
cloudrouter upload <id> ./my-project               # Upload directory to workspace
cloudrouter upload <id> ./config.json              # Upload single file to workspace
cloudrouter upload <id> . -r /home/user/app        # Upload to specific remote path
cloudrouter upload <id> . --watch                  # Watch and re-upload on changes
cloudrouter upload <id> . --delete                 # Delete remote files not present locally
cloudrouter upload <id> . -e "*.log"               # Exclude patterns
cloudrouter upload <id> . --dry-run                # Preview without making changes
 
# Download (sandbox -> local)
cloudrouter download <id>                          # Download workspace to current dir
cloudrouter download <id> ./output                 # Download workspace to ./output
cloudrouter download <id> ./output -r /home/user/app  # Download specific remote dir to ./output

This is the start of the file. Read the full SKILL.md on GitHub.

Security

Security scanPassed, 95 of 100Pattern checks on the skill's files.
Model reviewPassed

See the full security report

Quality evaluation

Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

See evaluation results

Source

SKILL.md in the repositoryCheckingskills/cloudrouter/SKILL.md

Popularity

7-day trend score: 0

Refreshed 23 Mar 2026, 7 months ago.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

Claude Code
Cursor
GitHub Copilot
Windsurf
Codex
Gemini CLI
Cline
Amp
Roo Code
Goose
Continue
Kilo Code
Junie
Trae
OpenHands
Qwen Code
AI Agent
Tabnine Chat
Sourcegraph Cody
Amazon Q
Firebase Genkit
Sweep
Devin
Phind
Mentat
Claude Desktop
Aider
GPT Pilot
Plandex
bolt.new
v0
Replit Agent
Anse
TypingMind
BetterChatGPT
LibreChat
Warp AI
Pieces
Fig AI

Embed badge

[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/karlorz/cloudrouter/cloudrouter/badge)](https://verified-skill.com/skills/karlorz/cloudrouter/cloudrouter)

Something wrong with this skill or its page? Tell us.

Install

SKILL.md

cloudrouter - Cloud Sandboxes for Development

When this skill is invoked

Setup

Quick Start

Commands

Authentication

Creating Sandboxes

Size Presets

GPU Options

All start Flags

Managing Sandboxes

Access Sandbox

Work with Sandbox

IMPORTANT: Fix npm Permissions Before Any npm Command

File Transfer

Security

Quality evaluation

Source

Popularity

Works with

All `start` Flags