mcp-server

Containarium

Name: Containarium
Author: FootprintAI

Open-source agent runtime — SSH-native isolation, eBPF egress policy, Kubernetes + LXC backends, GPU passthrough, MCP-native CLI

310 stars+5★ this week54 forksGoUpdated 10/22/2026100% free · open source

What it does

Containarium is an open-source agent runtime that provides SSH-native isolation, eBPF egress policy, and GPU passthrough for secure and efficient deployment of containers

When to use it

•When you need to deploy containers with SSH-native isolation for security
•When you require eBPF egress policy for network traffic control
•When you want to utilize GPU passthrough for compute-intensive workloads

Ready-to-paste prompt

./bin/containarium create --name=mygpucontainer --image=nvidia/cuda --gpu=0

Heads up: Ensure you have the necessary dependencies installed, including Go and the required Linux kernel version (5.10 or later) for eBPF support

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/containarium && curl -fsSL https://workflowstacks.com/api/skills/containarium/claude-skill -o ~/.claude/skills/containarium/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How Containarium works

Codeflow

Free to inspect

Containarium is a very large Go project (~625k lines across 2648 code files, plus 28 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit this month, Apache-2.0 license, has a test suite.

Size

Very large codebase

~625k lines · 2648 code files · days to read — use, don't read

Setup

Developer setup

A real software project. Use it via its install path; don't expect to read it all.

Runs on

No API keys detected

Go 90%Shell 4%TypeScript 4%HTML 1%

Where to start reading

1
README.md
Start here — what it does and how to install it
2
CLAUDE.md
The instructions the AI actually follows
3
cmd/agent-box/main.go
Where the program starts running
4
go.mod
Dependencies and the commands it exposes
5
internal/cmd/app.go
Inside internal/ — the main logic begins here

What's in each folder

internal/Core code — the actual logic2096 files

pkg/Core code — the actual logic393 files

cmd/Command-line entry points10 files

api/Backend / API1 files

docs/Documentation184 files

examples/Examples you can copy12 files

web-ui/Folder95 files

scripts/Helper scripts75 files

READMEHas testsDocumentedCI checksExamples includedApache-2.0 licenseUpdated this month

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/containarium && curl -fsSL https://workflowstacks.com/api/skills/containarium/claude-skill -o ~/.claude/skills/containarium/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

Quick Actions

Details

Creator

FootprintAI

Language