Responsible Vibe MCP

Appearance

Hands-On Tutorial ​

The best way to understand Responsible Vibe is to experience it. We'll build something simple, enhance it, then fix a bug. Three different workflows, three different approaches.

Prerequisites ​

Make sure you've got Responsible Vibe configured with your AI agent. If not, check the Quick Setup first.

Part 1: Greenfield Development ​

Let's build something from scratch. This triggers the Greenfield workflow – the full engineering process for new projects.

The Challenge ​

Ask your AI: "Build a simple dice game. Two players take turns rolling dice, highest total wins."

Hint: In order to not get just another Next.js app, you may instruct it to build for the terminal in a future stage

What You'll Experience ​

Your AI won't immediately start coding. Instead, it'll:

  1. Ask clarifying questions: "How many dice per roll? Best of how many rounds? Should we track scores across games?"

  2. Design the architecture: "I'm thinking a Game class, Player class, and a simple CLI interface. Does that sound right?"

  3. Plan the implementation: "Let me break this into phases: core game logic, player management, CLI interface, then testing."

  4. Build systematically: Following the plan it just created, not jumping around randomly.

This is the Greenfield workflow in action – comprehensive planning before implementation.

Key Observation ​

Notice how your AI is asking you questions instead of making assumptions. It's treating you as the product owner, not just someone who wants code written.

Part 2: Feature Enhancement ​

Now let's add something that wasn't in the original scope. This triggers the EPCC workflow (Explore → Plan → Code → Commit).

The Challenge ​

Ask your AI: "Add a high-score tracking system.
Players should see their win/loss record across games."

What You'll Experience ​

Different workflow, different approach:

  1. Explore: "Let me understand the current code structure and see how to integrate scoring..."

  2. Plan: "I'll need to modify the Player class, add persistent storage, and update the CLI to show stats."

  3. Code: Focused implementation of just the scoring feature

  4. Commit: Clean up and finalize the enhancement

This is EPCC – more iterative, focused on extending existing functionality rather than building from scratch.

Key Observation ​

The AI adapts its process based on context. It's not following the same heavy planning process as the greenfield project because it's working with existing code.

Part 3: Bug Fixing ​

Time to break something and fix it. This triggers the Bugfix workflow.

Create the Bug ​

First, let's introduce a bug manually:

  1. Find the dice rolling logic in your code
  2. Change something subtle (maybe make it always roll 1, or break the scoring)
  3. Save the file

The Challenge ​

Ask your AI: "There's a bug in the dice game.
Players are complaining the dice rolls seem unfair."

What You'll Experience ​

Yet another workflow approach:

  1. Reproduce: "Let me run the game and see if I can reproduce the issue..."

  2. Analyze: "I found the problem – the dice rolling logic is hardcoded to return 1."

  3. Fix: Targeted fix for just the bug, not a rewrite

  4. Verify: "Let me test this fix to make sure it works correctly..."

This is the Bugfix workflow – systematic debugging rather than random code changes.

Key Observation ​

The AI follows a methodical debugging process instead of just guessing at fixes. It reproduces first, then analyzes, then fixes.

What You Just Learned ​

You experienced three different engineering methodologies:

  • Greenfield: Comprehensive planning for new projects
  • EPCC: Iterative development for feature additions
  • Bugfix: Systematic debugging for problem resolution

Your AI automatically picked the right approach based on what you were trying to do. No configuration needed – it just works.

The Real Magic ​

This isn't just about following different steps. It's about your AI thinking like an engineer:

  • Asking the right questions at the right time
  • Planning before implementing when it matters
  • Adapting the process to the situation
  • Maintaining context across the entire development lifecycle

Most AI tools make you faster at writing code. Responsible Vibe makes your AI better at engineering software.

Next Steps ​

Try this tutorial with your own project ideas. The workflows adapt to any domain – web apps, CLI tools, games, whatever you're building.