Running the TeamBrain Agent with Claude CodeA Live Demo of Intent-Driven Engineering
- Mark Kendall
- 3 days ago
- 2 min read
Running the TeamBrain Agent with Claude Code
A Live Demo of Intent-Driven Engineering
Introduction
In this demonstration we will run the TeamBrain Agent directly from a repository using Claude Code inside Visual Studio Code.
The goal of this demo is to show how developers can move from intent → agent → running system in just a few steps.
Repository used in this demonstration:
This repository is part of the TeamBrain architecture, where a centralized “brain” repository governs rules and intelligence while service repositories execute those rules. (LearnTeachMaster Dev)
Demo Setup
Before running the agent, make sure you have:
• Git• Visual Studio Code• Claude Code installed in VS Code
Step 1 — Clone the Repository
Open a terminal and clone the repository.
git clone https://github.com/kendallmark3/teambrain-repro.git
Step 2 — Navigate to the Agent
Move into the TeamBrain agent directory.
cd teambrain-repro/teambrain-agent
This directory contains the agent configuration and intent definitions used to run the system.
Step 3 — Open the Repository in VS Code
Launch Visual Studio Code.
code .
Claude Code will now be able to inspect the repository and understand the project structure.
Step 4 — Ask Claude to Run the Agent
Open Claude Code inside VS Code and give it the following prompt.
You are inside the teambrain-agent repository.
Analyze the entire project.
Explain:
• the purpose of the agent
• the architecture
• how the agent should run
Then execute the steps required to start the agent locally.
Install any required dependencies and run the system.
Explain each step as you perform it.
Claude will now:
Read the repository
Understand the agent architecture
Install dependencies
Start the agent
Step 5 — Observe the Agent Running
Once Claude finishes executing the setup steps, the agent will start running locally.
You can now interact with the system and observe how the agent processes intent definitions and repository rules.
What This Demonstrates
This demo shows the power of Intent-Driven Engineering.
Instead of manually wiring together systems, developers can:
Define system intent
Provide guiding documentation
Allow an AI agent to interpret and run the system
The repository becomes a living specification, and the AI becomes the interpreter that brings that specification to life.
Related LearnTeachMaster Articles
Explore these articles for deeper explanations of the concepts demonstrated in this demo.
Team Brain Cognitive Governance Playbookhttps://www.learnteachmaster.org/post/the-team-brain-cognitive-governance-playbook
Intent-Driven Engineering with Claudehttps://www.learnteachmaster.org/post/intent-driven-engineering-with-claude
Claude Code and the Rise of Intent-Driven Developmenthttps://www.learnteachmaster.org/post/claude-code-and-the-rise-of-intent-driven-engineering
Bare-Bones Setup for Claude Codehttps://www.learnteachmaster.org/post/bare-bones-setup-for-claude-code-developer-quick-start
Why This Matters
The future of development is shifting toward intent-driven systems where:
• architecture is defined as intent• repositories become system specifications• AI agents implement and enforce those specifications
The TeamBrain approach demonstrates how this model can scale across many repositories using centralized governance. (LearnTeachMaster Dev)
If you'd like, I can also give you a much more powerful demo script that will make this video 10× more impressive:
It shows Claude doing full repository analysis + architectural diagram + running the agent automatically, which is a killer demo for your site.
Comments