# wetwork.dev

Many agents. One clean host. No loose ends.

wetwork.dev is an AI-agent development style built around one long-lived Docker host. The machine running the agent CLI stays clean: it edits files and controls Docker. The deliverable lives in `app/`; the agent workbench lives in `tools/`. Agents add narrow Dockerfiles and Compose services for tests, browsers, SDKs, scanners, migrations, and one-off jobs, then clean up the services they start.

## Example Repo Layout

```text
repo/
|-- AGENTS.md
|-- app/
|   |-- Dockerfile
|   `-- src/
`-- tools/
    |-- Dockerfile.test
    |-- Dockerfile.browser
    |-- docker-compose.yml
    `-- README.md
```

## AGENTS.md Files

The same methodology file is available for the major coding agents. Codex and Grok Build both use AGENTS.md:

```text
website/public/files/AGENTS.md
website/public/files/CLAUDE.md
website/public/files/GEMINI.md
```

Copy, curl, and wget commands are on the website:

```text
https://wetwork.dev/#rules
```

Full LLM-readable text:

```text
https://wetwork.dev/llms.txt
```

## Setup Instructions

Use a clean Ubuntu 26 LTS Docker host, including WSL. Install Docker first, add GPU container support only if the host needs it, then install the agent CLI directly on the host.

### 1. Install Docker Engine

#### Remove conflicting distro packages

Source: Docker Engine on Ubuntu - https://docs.docker.com/engine/install/ubuntu/

```bash
sudo apt remove $(dpkg --get-selections docker.io docker-compose docker-compose-v2 docker-doc podman-docker containerd runc | cut -f1)
```

#### Add Docker's official apt repository

Source: Docker Engine on Ubuntu - https://docs.docker.com/engine/install/ubuntu/

```bash
# Add Docker's official GPG key:
sudo apt update
sudo apt install ca-certificates curl
sudo install -m 0755 -d /etc/apt/keyrings
sudo curl -fsSL https://download.docker.com/linux/ubuntu/gpg -o /etc/apt/keyrings/docker.asc
sudo chmod a+r /etc/apt/keyrings/docker.asc

# Add the repository to Apt sources:
sudo tee /etc/apt/sources.list.d/docker.sources <<EOF
Types: deb
URIs: https://download.docker.com/linux/ubuntu
Suites: $(. /etc/os-release && echo "${UBUNTU_CODENAME:-$VERSION_CODENAME}")
Components: stable
Architectures: $(dpkg --print-architecture)
Signed-By: /etc/apt/keyrings/docker.asc
EOF

sudo apt update
```

#### Install Docker Engine, Buildx, and Compose plugin

Source: Docker Engine on Ubuntu - https://docs.docker.com/engine/install/ubuntu/

```bash
sudo apt install docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin
```

#### Verify Docker Engine

Source: Docker Engine on Ubuntu - https://docs.docker.com/engine/install/ubuntu/

```bash
sudo docker run hello-world
```

#### Allow your user to run Docker

Source: Docker Linux post-install - https://docs.docker.com/engine/install/linux-postinstall/

```bash
sudo groupadd docker
sudo usermod -aG docker $USER
newgrp docker
docker run hello-world
```

### 2. Optional NVIDIA GPU Container Support

Skip this section unless the host has NVIDIA GPUs that containers need to access.

#### Update before NVIDIA driver selection

Source: Ubuntu NVIDIA drivers - https://ubuntu.com/server/docs/how-to/graphics/install-nvidia-drivers/

```bash
sudo apt update && sudo apt upgrade
```

#### List GPU driver packages

Source: Ubuntu NVIDIA drivers - https://ubuntu.com/server/docs/how-to/graphics/install-nvidia-drivers/

```bash
sudo ubuntu-drivers list --gpgpu
```

#### Install Ubuntu's recommended compute driver

Source: Ubuntu NVIDIA drivers - https://ubuntu.com/server/docs/how-to/graphics/install-nvidia-drivers/

```bash
sudo ubuntu-drivers install --gpgpu
```

#### Install toolkit prerequisites

Source: NVIDIA Container Toolkit install - https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html

```bash
sudo apt-get update && sudo apt-get install -y --no-install-recommends \
   ca-certificates \
   curl \
   gnupg2
```

#### Add NVIDIA Container Toolkit apt repository

Source: NVIDIA Container Toolkit install - https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html

```bash
curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg \
  && curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list | \
    sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' | \
    sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list
```

#### Refresh apt after adding NVIDIA repository

Source: NVIDIA Container Toolkit install - https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html

```bash
sudo apt-get update
```

#### Install NVIDIA Container Toolkit packages

Source: NVIDIA Container Toolkit install - https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html

```bash
export NVIDIA_CONTAINER_TOOLKIT_VERSION=1.19.1-1
  sudo apt-get install -y \
      nvidia-container-toolkit=${NVIDIA_CONTAINER_TOOLKIT_VERSION} \
      nvidia-container-toolkit-base=${NVIDIA_CONTAINER_TOOLKIT_VERSION} \
      libnvidia-container-tools=${NVIDIA_CONTAINER_TOOLKIT_VERSION} \
      libnvidia-container1=${NVIDIA_CONTAINER_TOOLKIT_VERSION}
```

#### Configure Docker for NVIDIA runtime

Source: NVIDIA Container Toolkit install - https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html

```bash
sudo nvidia-ctk runtime configure --runtime=docker
```

#### Restart Docker after runtime configuration

Source: NVIDIA Container Toolkit install - https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html

```bash
sudo systemctl restart docker
```

#### Verify GPU access from Docker

Source: NVIDIA Container Toolkit sample workload - https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/sample-workload.html

```bash
sudo docker run --rm --runtime=nvidia --gpus all ubuntu nvidia-smi
```

### 3. Optional AMD ROCm Container Support

Skip this section unless the host has AMD GPUs that containers need to access.

#### Install AMD ROCm runtime

Source: Ubuntu 26 LTS release notes - https://documentation.ubuntu.com/release-notes/26.04/summary-for-lts-users/

```bash
apt install rocm
```

#### Install AMD ROCm development packages

Source: Ubuntu 26 LTS release notes - https://documentation.ubuntu.com/release-notes/26.04/summary-for-lts-users/

```bash
apt install rocm-dev
```

#### Configure ROCm shared objects

Source: ROCm post-install verification - https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/post-install.html

```bash
sudo tee --append /etc/ld.so.conf.d/rocm.conf <<EOF
/opt/rocm/lib
/opt/rocm/lib64
EOF
sudo ldconfig
```

#### Verify ROCm on the host

Source: ROCm post-install verification - https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/post-install.html

```bash
rocminfo | grep -i "Marketing Name:"
clinfo | grep -i "Board name:"
amd-smi version
```

#### Expose AMD GPUs to Docker containers

Source: ROCm Docker containers - https://rocm.docs.amd.com/projects/install-on-linux/en/latest/how-to/docker.html

```bash
docker run -it --rm \
    --device /dev/kfd \
    --device /dev/dri \
    --security-opt seccomp=unconfined \
    <image>
```

#### List AMD render nodes

Source: ROCm Docker containers - https://rocm.docs.amd.com/projects/install-on-linux/en/latest/how-to/docker.html

```bash
ls /dev/dri/render*
```

#### Run ROCm checks inside the container

Source: ROCm Docker containers - https://rocm.docs.amd.com/projects/install-on-linux/en/latest/how-to/docker.html

```bash
rocminfo
amd-smi list
```

### 4. Install Agent CLI

#### Codex CLI

Install Codex CLI with OpenAI's standalone installer.

##### Install Codex CLI

Source: OpenAI Codex CLI - https://github.com/openai/codex

```bash
curl -fsSL https://chatgpt.com/codex/install.sh | sh
```

##### Start Codex CLI

Source: OpenAI Codex CLI - https://github.com/openai/codex

```bash
codex
```

#### Grok Build

Install and start Grok Build from xAI's setup docs.

##### Install Grok Build

Source: Grok Build setup - https://docs.x.ai/build/overview

```bash
curl -fsSL https://x.ai/cli/install.sh | bash
```

##### Start Grok Build

Source: Grok Build setup - https://docs.x.ai/build/overview

```bash
grok
```

#### Claude Code

Install, start, and verify Claude Code from Anthropic's setup docs.

##### Install Claude Code

Source: Claude Code setup - https://docs.anthropic.com/en/docs/claude-code/setup

```bash
curl -fsSL https://claude.ai/install.sh | bash
```

##### Start Claude Code

Source: Claude Code setup - https://docs.anthropic.com/en/docs/claude-code/setup

```bash
claude
```

##### Verify Claude Code

Source: Claude Code setup - https://docs.anthropic.com/en/docs/claude-code/setup

```bash
claude doctor
```

#### Gemini CLI

Install and start Gemini CLI from Google's Gemini CLI docs.

##### Install nvm

Source: nvm install docs - https://github.com/nvm-sh/nvm

```bash
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.5/install.sh | bash
```

##### Load nvm in this shell

Source: nvm install docs - https://github.com/nvm-sh/nvm

```bash
export NVM_DIR="$([ -z "${XDG_CONFIG_HOME-}" ] && printf %s "${HOME}/.nvm" || printf %s "${XDG_CONFIG_HOME}/nvm")"
[ -s "$NVM_DIR/nvm.sh" ] && \. "$NVM_DIR/nvm.sh" # This loads nvm
```

##### Install current Node.js

Source: nvm install docs - https://github.com/nvm-sh/nvm

```bash
nvm install node # "node" is an alias for the latest version
```

##### Use current Node.js

Source: nvm install docs - https://github.com/nvm-sh/nvm

```bash
nvm use node
```

##### Install Gemini CLI

Source: Gemini CLI docs - https://google-gemini.github.io/gemini-cli/docs/get-started/

```bash
npm install -g @google/gemini-cli
```

##### Start Gemini CLI

Source: Gemini CLI docs - https://google-gemini.github.io/gemini-cli/docs/get-started/

```bash
gemini
```

## Template Repo

The downloadable template repo is generated from `website/server.py` and includes:

```text
AGENTS.md
CLAUDE.md
GEMINI.md
app/
tools/
docker-compose.yml
```

Download it from:

```text
https://wetwork.dev/download/wetwork.dev-template.tar.gz
```

## Website Source

This repository also contains the `wetwork.dev` website and Cloudflare Worker deploy tooling.

Local run:

```bash
docker compose -f website/docker-compose.yml up --build
```

```text
http://127.0.0.1:8787
```

Cloudflare deploy:

```bash
python3 cloudflare/deploy.py
```

Required local `.env` values:

```text
CLOUDFLARE_ACCOUNT_ID=...
CLOUDFLARE_API_TOKEN=...
```