Skip to content

schneidafunk/scrumteam

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Virtual Scrum Team Starter

A GitHub repository template featuring a virtual scrum team of AI agents that collaborate through text files and PRs. Perfect for learning Agile workflows, experimenting with AI-assisted development, or bootstrapping new projects with built-in quality gates.

Features

  • 6 AI Agent Personas — Product Owner, Scrum Master, Tech Lead, DevOps Engineer, QA Engineer, Security Engineer
  • Collaborative Review System — Agents communicate through markdown files and PRs
  • Human-AI Pairing — Join the virtual team as a contributor and collaborate with agents
  • Production-Ready Template — Community health files, CI/CD, security policies included
  • Fully Customizable — Modify personas, add new agents, or adapt for your workflow

Quick Start

Option 1: Use as Template (Recommended)

  1. Click "Use this template" on GitHub
  2. Clone your new repository
  3. Start collaborating with the virtual team!

Option 2: Fork & Clone

git clone https://github.com/YOUR_USERNAME/starter-copilot-repo.git
cd starter-copilot-repo

Meet the Virtual Scrum Team

Agent Role Specialty
@product-owner Product Owner Backlog prioritization, shipping decisions, ROI analysis
@scrum-master Scrum Master Facilitation, blocker removal, team health
@tech-lead Tech Lead Architecture, code quality, technical decisions
@devops-engineer DevOps Engineer CI/CD, infrastructure, reliability
@qa-engineer QA Engineer (SDET) Test automation, quality gates, bug detection
@security-engineer Security Engineer Vulnerability scanning, compliance, secure coding

How to Work with the Virtual Team

Step 1: Open a Discussion

Edit .github/scrum/discussions/current.md to start a new topic:

### Topic: Add user authentication feature

**Status**:  IN_PROGRESS

#### @human (You!)
> I'd like to add OAuth2 authentication. What should we consider?

#### @product-owner
> *Awaiting review...*

Step 2: Invoke an Agent

In VS Code with GitHub Copilot Chat, mention an agent:

@product-owner Review this feature request and provide acceptance criteria

Or ask for a full team review:

@scrum-master Facilitate a team discussion on the authentication feature

Step 3: Capture Agent Responses

Copy agent responses into the discussion file or their individual review files:

  • .github/scrum/reviews/product_owner_review.md
  • .github/scrum/reviews/tech_lead_review.md
  • .github/scrum/reviews/security_review.md

Step 4: Reach Consensus

The @scrum-master synthesizes feedback and the @product-owner makes the final call:

## Decision Queue

| Item | Proposed By | Status | Blocker |
|------|-------------|--------|---------|
| Add OAuth2 | @human |  APPROVED | None |

📁 Repository Structure

├── README.md                      # You are here!
├── LICENSE                        # MIT License
├── .gitignore                     # Comprehensive ignore patterns
├── requirements.txt               # Python dependencies (if needed)
│
├── .github/
│   ├── agents/                    #  Copilot agent definitions
│   │   ├── product_owner.agent.md
│   │   ├── scrum_master.agent.md
│   │   ├── tech_lead.agent.md
│   │   ├── devops_engineer.agent.md
│   │   ├── qa_engineer.agent.md
│   │   └── security_engineer.agent.md
│   │
│   ├── scrum/                     #  Team collaboration files
│   │   ├── discussions/
│   │   │   └── current.md         # Active sprint discussion
│   │   └── reviews/               # Individual agent reviews
│   │       ├── product_owner_review.md
│   │       ├── scrum_master_review.md
│   │       ├── tech_lead_review.md
│   │       ├── devops_review.md
│   │       ├── qa_review.md
│   │       └── security_review.md
│   │
│   ├── workflows/                 #  GitHub Actions
│   │   ├── ci.yml                 # Continuous integration
│   │   └── devops-agent.yml       # Weekly automated reviews
│   │
│   ├── CONTRIBUTING.md
│   ├── CODE_OF_CONDUCT.md
│   ├── SECURITY.md
│   └── copilot-instructions.md    # Copilot context for repo
│
└── scripts/
    ├── automation/
    │   └── devops_agent.py        # Automated maintenance script
    └── personas/                  #  Source of truth for personas
        ├── README.md
        ├── persona_product_owner.json
        ├── persona_scrum_master.json
        ├── persona_tech_lead.json
        ├── persona_devops_engineer.json
        ├── persona_qa_engineer.json
        └── persona_security_engineer.json

🔄 Example Workflow: Feature Development

A[Human creates feature request] --> B[@product-owner reviews value]
B --> C[@tech-lead reviews architecture]
C --> D[@security-engineer reviews risks]
D --> E[@qa-engineer defines test plan]
E --> F[@scrum-master synthesizes feedback]
F --> G[@product-owner approves/rejects]
G --> H[Human implements with @devops-engineer]
H --> I[PR created & merged]

💬 Sample Conversation

You (in Copilot Chat):

@tech-lead I'm considering using JWT for authentication. 
What are the trade-offs vs session-based auth?

@tech-lead responds:

 ARCHITECTURE REVIEW:

JWT vs Session-Based Authentication:

| Aspect | JWT | Sessions |
|--------|-----|----------|
| Scalability |  Stateless |  Requires session store |
| Security |  Can't revoke easily |  Server-controlled |
| Complexity |  Token refresh logic |  Simpler |

 RECOMMENDATION: For a new project, start with sessions 
unless you need microservices or mobile apps.

❓ QUESTION FOR @security-engineer: What's our token 
expiry policy if we go with JWT?

Customization

Add a New Agent

  1. Create persona in scripts/personas/persona_<role>.json
  2. Create agent in .github/agents/<role>.agent.md
  3. Add review file in .github/scrum/reviews/<role>_review.md
  4. Update .github/copilot-instructions.md with the new role

Modify Review Cadence

Edit .github/workflows/devops-agent.yml to change the schedule:

on:
  schedule:
    - cron: '0 3 * * 1'  # Every Monday at 03:00 UTC

📚 Documentation

Document Purpose
Contributing Guide How to contribute
Code of Conduct Community standards
Security Policy Vulnerability reporting
Personas README Persona schema & details

🤝 Contributing

  1. Fork the repository
  2. Create a feature branch (git checkout -b feature/amazing-feature)
  3. Start a discussion with the virtual team
  4. Implement changes based on team feedback
  5. Open a PR and let agents review!

📄 License

MIT License — see LICENSE for details.

🙋 FAQ

Q: Do I need GitHub Copilot to use this?
A: The agents work best with GitHub Copilot Chat, but you can manually role-play the personas using the JSON definitions.

Q: Can I use this for a real project?
A: Absolutely! Add your source code to src/ and customize the agents for your domain.

Q: How do I add my own persona?
A: Copy an existing persona JSON, modify it, and create a matching agent.md file.


Built with for humans and AI working together

About

No description, website, or topics provided.

Resources

License

Code of conduct

Contributing

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published