modules/KENL3: Development Environments

Icon: 💻 | Color: Green | Status: Production Ready

Containerized development environments with distrobox, Claude Code integration, MCP servers, and local AI.

Quick Start

# Setup development environment
cd ~/kenl/KENL3-dev
./setup-devenv.sh ubuntu  # or fedora, debian

# Switch to dev context
cd ~/kenl/KENL5-facades
./switch-kenl.sh dev

What changes:

stateDiagram-v2
    [*] --> HostBazzite
    HostBazzite --> DevContainer: distrobox create
    DevContainer --> ClaudeCode: Claude Code setup
    ClaudeCode --> MCPServers: MCP integration
    MCPServers --> LocalAI: Ollama + Qwen

Before	After
Host-only development	Isolated dev containers
No AI assistance	Claude Code + Local AI
Manual configuration	ATOM-tracked setups

Features

🐧 Distrobox Environments: Ubuntu 24.04, Fedora 41, Debian stable
🤖 Claude Code Integration: MCP servers, custom prompts
🧠 Local AI: Ollama with Qwen 2.5 models
📦 Devcontainers: VS Code/Claude Code .devcontainer configs
🔌 MCP Servers: Cloudflare, filesystem, GitHub integrations
📚 System Documentation: Project-specific docs and context

Directory Structure

KENL3-dev/
├── distrobox-envs/        # Distrobox environment configs
│   ├── ubuntu24-dev.ini   # Ubuntu 24.04 LTS
│   ├── fedora41-dev.ini   # Fedora 41
│   └── debian-stable.ini  # Debian stable
├── devcontainer/          # Devcontainer configurations
│   └── bazzite-example/   # Example Bazzite devcontainer
├── claude-code-setup/     # Claude Code configuration
│   └── claude-configuration-guide.md
├── mcp-configs/           # MCP server configurations
│   ├── cloudflare.json    # Cloudflare MCP
│   ├── filesystem.json    # Local filesystem MCP
│   └── github.json        # GitHub MCP
├── local-ai/              # Local AI setup (Ollama + Qwen)
│   ├── install-ollama.sh
│   ├── models/            # Qwen model configs
│   └── prompts/           # Custom prompts
├── system-docs/           # System-specific documentation
│   └── .kenl/             # modules/KENL system context docs
└── setup-devenv.sh        # Main setup script

Development Environments

Ubuntu 24.04 LTS (Recommended)

# Create Ubuntu dev environment
./setup-devenv.sh ubuntu

# What it installs:
# - Python 3.12 + pip + venv
# - Node.js 20 LTS + npm
# - Git, curl, wget, build-essential
# - Docker CLI (for docker-compose)
# - Claude Code CLI
# - ATOM framework integration

Use for:

Python projects (Django, Flask, FastAPI)
Node.js/TypeScript development
General-purpose development
Claude Code primary environment

Fedora 41 (Cutting Edge)

# Create Fedora dev environment
./setup-devenv.sh fedora

# What it installs:
# - Latest Podman, buildah, skopeo
# - Rust toolchain (rustc, cargo)
# - Go 1.21+
# - Development tools (gcc, clang, cmake)

Use for:

Container development
Rust/Go projects
System programming
Testing latest tools

Debian Stable (Ultra Stable)

# Create Debian dev environment
./setup-devenv.sh debian

# What it installs:
# - Long-term stable packages
# - Conservative toolchain
# - Minimal dependencies

Use for:

Production-like environments
Long-term projects
Security-sensitive work

Claude Code Integration

Initial Setup

cd ~/kenl/KENL3-dev/claude-code-setup

# Install Claude Code CLI
curl -fsSL https://claude.ai/install.sh | sh

# Configure MCP servers
cp mcp-configs/*.json ~/.config/claude/

What changes:

~/.config/claude/
+ ├── mcp-servers.json    # MCP server registry
+ ├── prompts/            # Custom prompts
+ └── settings.json       # Claude Code settings

MCP Server Configuration

Cloudflare MCP (mcp-configs/cloudflare.json):

{
  "mcpServers": {
    "cloudflare": {
      "command": "npx",
      "args": ["-y", "@cloudflare/mcp-server-cloudflare"],
      "env": {
        "CLOUDFLARE_API_TOKEN": "${CLOUDFLARE_API_TOKEN}",
        "CLOUDFLARE_ACCOUNT_ID": "${CLOUDFLARE_ACCOUNT_ID}"
      }
    }
  }
}

GitHub MCP (mcp-configs/github.json):

{
  "mcpServers": {
    "github": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-github"],
      "env": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "${GITHUB_TOKEN}"
      }
    }
  }
}

Filesystem MCP (mcp-configs/filesystem.json):

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "/home/user"],
      "env": {}
    }
  }
}

Local AI Setup (Ollama + Qwen)

Installation

cd ~/kenl/KENL3-dev/local-ai
./install-ollama.sh

# Download Qwen 2.5 models
ollama pull qwen2.5:7b       # 7B model (4GB RAM)
ollama pull qwen2.5:14b      # 14B model (8GB RAM)
ollama pull qwen2.5-coder:7b # Code-specialized

Storage location: /mnt/claude-ai/ollama (external drive, modules/KENL9)

What changes:

flowchart LR
    A[Install Ollama] --> B[Download Qwen models]
    B --> C[Configure API endpoint]
    C --> D[Integrate with Claude Code]
    D --> E[✅ Local AI ready]

Model Selection

Model	Size	RAM	Use Case
qwen2.5:7b	4.7GB	8GB	General assistance
qwen2.5:14b	8.6GB	16GB	Complex reasoning
qwen2.5-coder:7b	4.7GB	8GB	Code generation
qwen2.5:32b	19GB	32GB	Maximum capability

Usage Example

# Direct Ollama usage
ollama run qwen2.5-coder:7b

# Via API (for Claude Code integration)
curl http://localhost:11434/api/generate -d '{
  "model": "qwen2.5-coder:7b",
  "prompt": "Write a Python function to parse YAML",
  "stream": false
}'

Distrobox Workflows

Creating Named Environments

# Create project-specific environment
distrobox create \
  --name bazza-dev \
  --image docker.io/library/ubuntu:24.04 \
  --home ~/distrobox/bazza-dev \
  --init

# Enter environment
distrobox enter bazza-dev

# Inside container, install modules/KENL framework
cd ~/kenl/KENL1-framework/atom-sage-framework
./install.sh

Exporting Applications

# Export VS Code from container to host
distrobox-export --app code

# Now "code" command works from host, runs in container
# Icon appears in application menu

Sharing Directories

# Host directories auto-mounted in container:
# - $HOME → /home/user
# - /mnt → /mnt (game libraries, AI data)
# - /var/home → /var/home

# Containers can access all modules/KENL modules directly!

Devcontainer Configurations

Bazzite Development Container

See devcontainer/bazzite-example/.devcontainer.json:

{
  "name": "Bazzite Development",
  "image": "ghcr.io/ublue-os/bazzite-arch:latest",
  "features": {
    "ghcr.io/devcontainers/features/common-utils:2": {},
    "ghcr.io/devcontainers/features/git:1": {}
  },
  "postCreateCommand": "bash .devcontainer/post-create.sh",
  "customizations": {
    "vscode": {
      "extensions": [
        "anthropics.claude-code",
        "ms-python.python",
        "ms-vscode.cpptools"
      ]
    }
  }
}

Post-create script (post-create.sh):

Installs modules/KENL1 framework
Configures ATOM trail
Sets up MCP servers
Installs local AI (optional)

System Documentation

The system-docs/.kenl/ directory contains host-specific documentation:

host-system-issues.md: Known hardware/software issues
gpu-hang-warning.md: NVIDIA GPU hang workarounds
REBASE_EXPECTATIONS.md: rpm-ostree rebase guide
CURRENT_VS_POST_REBASE.md: Version comparison

These docs travel with backups (KENL10) for disaster recovery context!

AI Integration Levels

(See also: AI-INTEGRATION-GUIDE.md)

modules/KENL3 AI Usage: 🟩 Maximum Help

What AI helps with:

Code generation and refactoring
Debugging and error resolution
Documentation writing
Test creation
Dependency management
Architecture design

Recommended AI stack:

Claude Code (primary): Real-time coding assistance
Qwen 2.5 Coder (local): Privacy-sensitive code, offline work
Perplexity (research): API documentation, library selection

Example workflow:

# Start Claude Code in dev environment
distrobox enter bazza-dev
cd ~/projects/my-app

claude code .

# Claude can now:
# - Access project via filesystem MCP
# - Query GitHub repos via GitHub MCP
# - Deploy to Cloudflare via Cloudflare MCP
# - Use local Qwen for code completion

Dependencies

Requires:

KENL0 (system): Podman/distrobox installation
KENL1 (framework): ATOM trail for environment tracking

Used by:

KENL2 (gaming): Play Card creation (research + AI)
KENL4 (monitoring): Dashboard development
KENL11 (media): Docker compose configuration

Common Workflows

1. Create New Project Environment

# Create and enter environment
./setup-devenv.sh ubuntu my-project

# Install modules/KENL framework
cd ~/kenl/KENL1-framework/atom-sage-framework && ./install.sh

# Track environment creation
atom CONFIG "Created dev environment for my-project" "distrobox create my-project"

2. Set Up Claude Code with MCP

# Copy MCP configs
cp mcp-configs/*.json ~/.config/claude/

# Set environment variables
echo 'export CLOUDFLARE_API_TOKEN="..."' >> ~/.bashrc
echo 'export GITHUB_TOKEN="..."' >> ~/.bashrc
source ~/.bashrc

# Test MCP servers
claude mcp list

3. Install Local AI for Offline Development

# Install Ollama
cd local-ai && ./install-ollama.sh

# Download code model
ollama pull qwen2.5-coder:7b

# Test
ollama run qwen2.5-coder:7b "Write a hello world in Rust"

4. Export Container App to Host

# Inside container, install app
sudo apt install code

# Export to host
distrobox-export --app code

# Now "code" works from host launcher!

Troubleshooting

distrobox: command not found

# Install on Bazzite
rpm-ostree install distrobox
systemctl reboot

# Or use toolbox (pre-installed)
toolbox create bazza-dev
toolbox enter bazza-dev

MCP server not found

# Ensure Node.js available
node --version  # Should be 18+

# Install MCP server manually
npm install -g @cloudflare/mcp-server-cloudflare

# Verify
which mcp-server-cloudflare

Ollama models slow

# Check GPU usage
nvidia-smi  # For NVIDIA
radeontop   # For AMD

# Enable GPU passthrough in container
distrobox create \
  --nvidia  # For NVIDIA
  # OR
  --additional-flags "--device=/dev/dri" # For AMD

Container can't access /mnt

# Ensure mounts configured
distrobox create \
  --additional-flags "--volume=/mnt:/mnt:rslave"

Navigation

← Root README: Overview of all modules/KENL modules
→ KENL1: Framework: Core ATOM/SAGE framework
→ KENL4: Monitoring: Dashboard development
→ AI Integration Guide: Complete AI setup guide

Status: Production Ready | Icon: 💻 | AI Level: 🟩 Maximum Help

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

modules/KENL3: Development Environments

Quick Start

Features

Directory Structure

Development Environments

Ubuntu 24.04 LTS (Recommended)

Fedora 41 (Cutting Edge)

Debian Stable (Ultra Stable)

Claude Code Integration

Initial Setup

MCP Server Configuration

Local AI Setup (Ollama + Qwen)

Installation

Model Selection

Usage Example

Distrobox Workflows

Creating Named Environments

Exporting Applications

Sharing Directories

Devcontainer Configurations

Bazzite Development Container

System Documentation

AI Integration Levels

modules/KENL3 AI Usage: 🟩 Maximum Help

Dependencies

Common Workflows

1. Create New Project Environment

2. Set Up Claude Code with MCP

3. Install Local AI for Offline Development

4. Export Container App to Host

Troubleshooting

distrobox: command not found

MCP server not found

Ollama models slow

Container can't access /mnt

Navigation

FilesExpand file tree

README.md

Latest commit

History

README.md

File metadata and controls

modules/KENL3: Development Environments

Quick Start

Features

Directory Structure

Development Environments

Ubuntu 24.04 LTS (Recommended)

Fedora 41 (Cutting Edge)

Debian Stable (Ultra Stable)

Claude Code Integration

Initial Setup

MCP Server Configuration

Local AI Setup (Ollama + Qwen)

Installation

Model Selection

Usage Example

Distrobox Workflows

Creating Named Environments

Exporting Applications

Sharing Directories

Devcontainer Configurations

Bazzite Development Container

System Documentation

AI Integration Levels

modules/KENL3 AI Usage: 🟩 Maximum Help

Dependencies

Common Workflows

1. Create New Project Environment

2. Set Up Claude Code with MCP

3. Install Local AI for Offline Development

4. Export Container App to Host

Troubleshooting

distrobox: command not found

MCP server not found

Ollama models slow

Container can't access /mnt

Navigation