Setup Guide — Vibe Coding for Localization Engineers

What you'll build

A TMX ↔ XLIFF converter

A tool localization professionals use every day — one that normally requires expensive specialised software. You'll leave with working code you can run and extend at your own pace. Tasks are scoped small so everyone can complete them in the session.

Before the workshop

⚠

Required — please complete before May 20

We won't have time to troubleshoot installations during the session. This takes roughly 10 minutes. Complete it in advance so you can start building from minute one.

Install Python 3

The converter we'll build during the workshop runs on Python 3. Check if you already have it — if yes, skip this step entirely.

1
Check if Python is already installed:
Terminal · macOS
```
python3 --version
```
If you see Python 3.x.x — you're done.
2

If not installed — go to python.org/downloads and click the big yellow Download Python 3.x.x button. Open the downloaded file and click through the installer.

xmllint — validates that TMX and XLIFF files are well-formed. Already installed on macOS. Verify:

Terminal · macOS

xmllint --version

If it's missing, run xcode-select --install and it will appear.

1
Check if Python is already installed:
PowerShell · Windows
```
python --version
```
If you see Python 3.x.x — you're done.
2
If not installed — the fastest way is one command in PowerShell:
PowerShell · Windows
```
winget install Python.Python.3
```
Or go to python.org/downloads, download the installer, and run it. Important: on the first installer screen, tick "Add Python to PATH" before clicking Install.

xmllint — validates TMX and XLIFF files. Install with one command:

PowerShell · Windows

winget install xmlsoft.libxml2

After installing, close and reopen PowerShell, then verify with xmllint --version.

Install OpenClaw

OpenClaw is the tool we'll use throughout the session. Select your OS below and run the command in your terminal.

docs.openclaw.ai/install

1

Open Terminal — see how below if you're not sure
2
Paste and run this command:
Terminal · macOS
```
curl -fsSL https://openclaw.ai/install.sh | bash
```
Official installer — detects your OS, installs Node if needed, installs OpenClaw, and launches the setup wizard.
3

Follow the on-screen prompts to finish the wizard.

How to open Terminal macOS

Press ⌘ Command + Space to open Spotlight
Type Terminal and press Return
Paste the command above and press Return

Using an AI coding assistant?

If you have Claude Code, GitHub Copilot, or GitHub Codex — paste the install command directly and ask it to guide you through setup. Say: "Help me run this OpenClaw install command."

1

Open PowerShell — see how below
2
Paste and run this command:
PowerShell · Windows
```
& ([scriptblock]::Create((iwr -useb https://openclaw.ai/install.ps1)))
```
Official Windows installer. PowerShell works well; WSL2 also supported for extra stability.
3

Follow the on-screen prompts to finish the wizard.

How to open PowerShell Windows

Option A: Press Win + X → click Windows PowerShell or Windows Terminal
Option B: Press Win + R, type powershell, press Enter
Option C: Start menu → search PowerShell → open

Using an AI coding assistant?

If you have Claude Code, GitHub Copilot, or GitHub Codex — paste the install command directly and ask it to guide you through setup. Say: "Help me run this OpenClaw install command."

Configure OpenRouter

OpenClaw uses OpenRouter to access AI models. Connect your API key so the workshop tools can run. Custom.MT will provide the key before the session.

docs.openclaw.ai — OpenRouter provider openrouter.ai — OpenClaw integration guide

Easiest: use the guided wizard

The onboard command walks you through the entire OpenRouter setup interactively. Recommended if this is your first time.

Terminal · macOS & Windows

openclaw onboard

When you see the prompt ◆ Model/auth provider, scroll to openrouter in the list and press Enter. Then paste the API key from Custom.MT (begins with sk-or-...).

Or set the key directly without the wizard:

1
Open Terminal (same as Step 1)
2
Set your OpenRouter key:
Terminal · macOS
```
openclaw auth set openrouter:default --key "YOUR_OPENROUTER_KEY"
```
Replace YOUR_OPENROUTER_KEY with the key from Custom.MT (begins with sk-or-...).
3
Verify the key was saved:
Terminal · macOS
```
openclaw auth list
```
You should see openrouter:default listed as a configured provider.

Or set the key directly without the wizard:

1
Open PowerShell (same as Step 1)
2
Set your OpenRouter key:
PowerShell · Windows
```
openclaw auth set openrouter:default --key "YOUR_OPENROUTER_KEY"
```
Replace YOUR_OPENROUTER_KEY with the key from Custom.MT (begins with sk-or-...).
3
Verify the key was saved:
PowerShell · Windows
```
openclaw auth list
```
You should see openrouter:default listed as a configured provider.

Set the model & verify

Configure OpenClaw to use Kimi K2.6 — the model we'll use in the workshop — then launch the WebUI and confirm it responds in the chat. If you get an answer, you're fully set up.

openrouter.ai — Kimi K2.6 model page

1

Open Terminal / PowerShell (same as Steps 1 & 2)
2
Set Kimi K2.6 as your active model:
Terminal · macOS & Windows
```
openclaw models set openrouter/moonshotai/kimi-k2.6
```
If prompted to choose from a list, select "Enter manually" and type the full model ID: openrouter/moonshotai/kimi-k2.6

When prompted "How do you want to hatch your bot?" — choose Open the Web UI:

Prompt

How do you want to hatch your bot?
│  ○ Hatch in Terminal (recommended)
│  ● Open the Web UI
│  ○ Do this later

4
Or launch the WebUI at any time with:
Terminal · macOS & Windows
```
openclaw dashboard
```
This opens the browser dashboard at http://127.0.0.1:18789. If it doesn't open automatically, paste that address into your browser.
5

Send a test message in the chat — type anything, for example: "Hello, what model are you?"

If Kimi K2.6 responds in the chat window, your setup is complete. ✓

No response in the chat?

Run the doctor command — it detects common issues and can fix them automatically:

Terminal · macOS & Windows

openclaw doctor --fix

Also check that openclaw auth list shows openrouter:default, and that the model name is spelled exactly as shown above — including the openrouter/ prefix.

Pre-workshop checklist

Confirm all four before May 20:

Python 3 installed — python3 --version (macOS) or python --version (Windows) prints a version number

xmllint installed — xmllint --version prints a version number

OpenClaw installed — running openclaw --version prints a version number

OpenRouter key configured — no error when running the auth command

Text editor ready — VS Code, Notepad, TextEdit, any application works

Need help before the session?

Reach out to the Custom.MT team if you run into issues. The OpenClaw FAQ also covers the most common installation problems.