# Sandboxes Cli

> Manage Koyeb sandboxes: create, list, run commands, manage processes, filesystem operations, and port exposure using the CLI.

- **Type:** Skill
- **Install:** `agentstack add skill-koyeb-koyeb-skills-sandboxes-cli`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [koyeb](https://agentstack.voostack.com/s/koyeb)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [koyeb](https://github.com/koyeb)
- **Source:** https://github.com/koyeb/koyeb-skills/tree/main/plugins/koyeb/skills/sandboxes-cli

## Install

```sh
agentstack add skill-koyeb-koyeb-skills-sandboxes-cli
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Koyeb Sandboxes (CLI)

## When to use

Use this skill to interact with Koyeb sandboxes created via `koyeb service create --type=sandbox`, including running commands, managing processes, files, and ports.

## Prerequisites

- Koyeb CLI installed and on PATH
- Authenticated session (see references/koyeb-auth.md)

## Workflow

1. Create or identify a sandbox.
2. Check health and list running processes.
3. Run commands or start background processes.
4. Expose or unexpose ports as needed.
5. Use filesystem commands for data inspection.

## Commands

- Create sandbox: `koyeb sandbox create SANDBOXNAME --app APPNAME [flags]`
- List sandboxes: `koyeb sandbox list SANDBOXNAME/APPNAME[flags]`
- Check health: `koyeb sandbox health SANDBOXNAME/APPNAME [flags]`
- Run command: `koyeb sandbox run SANDBOXNAME/APPNAME --  [args...]`
- Start process: `koyeb sandbox start SANDBOXNAME/APPNAME --  [args...]`
- List processes: `koyeb sandbox ps SANDBOXNAME/APPNAME [flags]`
- Kill process: `koyeb sandbox kill SANDBOXNAME/APPNAME  [flags]`
- Stream logs: `koyeb sandbox logs SANDBOXNAME/APPNAME  [flags]`
- Filesystem operations: `koyeb sandbox fs SANDBOXNAME/APPNAME  [flags]`
- Expose port: `koyeb sandbox expose-port SANDBOXNAME/APPNAME  [flags]`
- Unexpose port: `koyeb sandbox unexpose-port SANDBOXNAME/APPNAME [flags]`

## Examples

- Create a sandbox: `koyeb sandbox create --name my-sandbox`
- Run a command: `koyeb sandbox run SANDBOXNAME/APPNAME -- ls -la`
- Start a background process: `koyeb sandbox start SANDBOXNAME/APPNAME -- python app.py`
- Expose a port: `koyeb sandbox expose-port SANDBOXNAME/APPNAME 8080`

## References

For detailed flags, see references/koyeb-sandbox-flags.md.

- references/koyeb-cli.md
- references/koyeb-sandbox-flags.md
- references/koyeb-auth.md
- references/koyeb-output.md

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [koyeb](https://github.com/koyeb)
- **Source:** [koyeb/koyeb-skills](https://github.com/koyeb/koyeb-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-koyeb-koyeb-skills-sandboxes-cli
- Seller: https://agentstack.voostack.com/s/koyeb
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.