Skip to Content
Getting StartedFor AI-First Builders

Getting Started for AI-First Builders

You’re building with AI as your primary collaborator — writing features through conversation, not just autocomplete. Dot•requirements helps you maintain coherence as your project grows beyond a prototype.

Why Capture Requirements?

“Won’t this slow me down?” No — it speeds you up.

When you describe what you’re building before the AI writes code:

  • AI gets it right the first time — clear requirements mean fewer rewrites
  • Day-two features have a foundation — your AI assistant can read what came before
  • *Linking tests to requirements keeps AI honest — you can trust that your tests are actually validating the behaviour you asked for

Like a well-crafted prompt, a lightweight spec takes a few extra minutes upfront but saves hours of debugging imprecise assumptions later.

Also, for when you’d rather jump in and prototype something, build-specify-test is a totally valid workflow — you come away with the same strong foundation for future work.

Set Up MCP

MCP (Model Context Protocol) lets your AI assistant use dot•requirements tools directly. Setup takes about a minute.

1. Install the CLI

npm install -g @popoverai/dotrequirements

2. Log in to dot•requirements cloud

dotreq login

This opens a browser to authenticate. A cloud account is optional, but it enables style checking, test review, and coverage sync (the features that make AI-assisted development actually work well).

3. Initialize your project

Creates your project and sets up credentials in .env.local.

4. Configure your AI assistant

dotreq mcp-setup

This interactive command:

  • Asks which AI assistant you use
  • Configures the MCP server automatically
  • Offers to install workflow skills (recommended)

Supported assistants:

  • Claude Code
  • Claude Desktop
  • Cursor
  • Google Antigravity
  • OpenAI Codex
  • GitHub Copilot

After setup, restart your AI assistant for changes to take effect.

The Workflow

Once MCP is configured, your AI assistant can use dot•requirements tools. Here’s how a typical feature development looks:

You describe what you want to build

“Add a login page with email and password. Users should see an error if credentials are wrong, and get redirected to the dashboard on success.”

AI captures requirements

Your assistant will ask: “Should we capture what this should do before we build it?”

If you say yes, it:

  1. Creates a requirements document with clear, testable statements
  2. Runs style_check to catch clarity issues
  3. Shows you the requirements for review

You review and approve

Check that the requirements capture your intent. Adjust if needed. The AI pushes approved requirements to the cloud with push_requirements.

AI implements and tests

Your assistant writes the code, then writes tests that reference the requirements.

After writing tests, the AI runs review_test to verify tests actually validate what the requirements specify — not just that they pass.

You accept the result

Run the tests, verify the implementation works, and move on. Coverage is tracked automatically.

Next Steps

Last updated on
For Product ManagersHow It Works