intellimetrics Learning
Applied The OFCI Playbook

The Procurement Workbench

Last reviewed · content updated

Beginner

What you'll learn

~15 min
  • Open a terminal on your operating system
  • Install Node.js using nvm
  • Install and authenticate an AI CLI tool (or use browser chat as an alternative)
  • Verify everything works with a test command

Before you build the RFP Machine

Every tool in this module runs from the command line. Before you start generating RFPs, scoring vendors, or tracking submittals, you need three things on your machine:

  1. A terminal — the app where you type commands
  2. Node.js — the runtime that powers the tools
  3. An AI CLI tool — the engine that builds everything

Fifteen minutes from now, you’ll be ready to go. This is equipping your workbench, not sitting through a lecture.

💡Don't want to install anything?

You can use Claude.ai, Gemini, or ChatGPT in your browser instead. Paste the same prompts from the lessons, get the same tools generated, then download the files and run them locally later. The browser path works fine — you just won’t get the “type a command and watch it build” experience until you set up the CLI. Skip to Verify it works if you’re going browser-first.

Step 1: Open a terminal (2 minutes)

The terminal is where you’ll run every tool in this module. It’s just a text interface to your computer — nothing scary, nothing breakable.

Mac:

  1. Press Cmd + Space to open Spotlight
  2. Type Terminal and press Enter
  3. You’ll see a window with a blinking cursor — that’s it, you’re in

Windows:

  1. Open the Start menu and search for Ubuntu or WSL
  2. Click the Ubuntu app to open a Linux terminal inside Windows
  3. If you don’t have WSL installed: open PowerShell as Administrator, run wsl --install, restart, then open Ubuntu

Linux:

  1. Press Ctrl + Alt + T to open a terminal

You should see a prompt that looks something like user@machine:~$ — that’s your command line, ready for input.

ℹWant the full version?

The shared curriculum covers terminals in depth: What Is a Terminal?, Opening Your Terminal, and Navigating the Filesystem. Come back to those if you want a deeper understanding later.

Step 2: Install Node.js via nvm (5 minutes)

Node.js is the runtime that powers the CLI tools you’ll build. The easiest way to install it is through nvm (Node Version Manager) — it handles everything for you.

Install nvm:

Terminal window
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.3/install.sh | bash

Close and reopen your terminal, then:

Terminal window
nvm install --lts

Verify it worked:

Terminal window
node --version
npm --version

You should see version numbers (e.g., v22.x.x and 10.x.x). If you do, Node.js is ready.

ℹWant the full version?

The shared curriculum covers Node.js setup in detail: Overview & Prerequisites. It includes proxy/certificate troubleshooting for corporate networks and alternative install methods.

Step 3: Install an AI CLI tool (5 minutes)

You need one of these. Pick whichever fits your situation — they all work with the prompts in this module.

Terminal window
npm install -g @anthropic-ai/claude-code
claude

Follow the authentication prompts. Once you see the Claude Code interface, you’re ready.

Option B: Antigravity CLI (Google)

Google retired the consumer Gemini CLI on 2026-06-18. Individuals now install Antigravity CLI (agy):

Terminal window
curl -fsSL https://antigravity.google/cli/install.sh | bash
agy

On an enterprise Gemini Code Assist license, Gemini CLI still works — see Gemini CLI → Antigravity CLI for both paths.

Option C: Codex CLI

Terminal window
npm install -g @openai/codex
codex
💡Investment framing

Claude Code costs ~$20/month through a Pro subscription. GitHub Copilot has a free tier (Copilot Free). Think of it like any other professional tool subscription — if it saves you two hours on one RFP cycle, it’s paid for itself in the first week.

ℹWant the full version?

The shared curriculum covers each tool’s setup in detail: Claude Code, Antigravity, Codex CLI. Those lessons include SSH authentication, API key setup, and troubleshooting for restricted networks.

Verify it works

Run this quick test to confirm everything is connected:

If you installed a CLI tool:

Terminal window
mkdir test-workbench && cd test-workbench
claude "Create a file called hello.js that prints 'Workbench ready' to the console"
node hello.js

You should see Workbench ready printed. If you do — you’re set. Delete the test folder and move on to Lesson 1.

Terminal window
cd ..
rm -rf test-workbench

If you’re going browser-first:

  1. Open claude.ai (or your preferred AI chat)
  2. Paste: Write a Node.js file called hello.js that prints "Workbench ready" to the console
  3. You’ll see the generated code — that’s the same thing the CLI produces, just in a browser window

Either path works. The CLI is faster and more powerful for file-heavy work, but browser chat gets you started with zero installation.

You’re ready

Your workbench is equipped. In the next lesson, you’ll use these tools to build The RFP Machine — a CLI that turns rough equipment specs into complete, CoreSite-formatted RFP documents in 30 seconds.

Time to stop configuring and start building.

Search lessons