Skip to content

quick-start.md

Latest commit

 

History

History
236 lines (156 loc) · 5.51 KB

quick-start.md

File metadata and controls

236 lines (156 loc) · 5.51 KB
Raw

🚀 Quick Start Guide

Get AGOR running in 5 minutes and experience the power of coordinated AI development.

🎯 What You'll Accomplish

By the end of this guide, you'll have:

  • ✅ AGOR installed and working
  • ✅ Your first project bundled
  • ✅ An AI agent coordinating your development workflow
  • ✅ Understanding of the core concepts

🛠️ Prerequisites

  • Python 3.10+ installed
  • A development project you want to coordinate
  • AI platform access (Google AI Studio recommended)

📦 Step 1: Install AGOR

# Install AGOR (recommended method)
pipx install agor

# Verify installation
agor --version
# Should output: agor, version 0.4.1

Don't have pipx? Install it first: pip install pipx

🎁 Step 2: Bundle Your Project

# Navigate to your project
cd /path/to/your/project

# Bundle for Google AI Studio
agor bundle . -f zip

# Or bundle for ChatGPT
agor bundle .



# AGOR creates: your-project-agor-bundle.zip (or .tar.gz)

What just happened?

  • AGOR analyzed your project structure
  • Packaged your code with git history
  • Included a portable git binary
  • Generated AI instructions
  • Created an upload-ready bundle

🤖 Step 3: Choose Your AI Platform

Option A: Google AI Studio

  1. Visit Google AI Studio
  2. Create new chat and enable Function Calling ⚠️
  3. Upload your .zip bundle
  4. Copy-paste the generated prompt

Option B: ChatGPT (Subscription)

  1. Open ChatGPT
  2. Upload your .tar.gz bundle
  3. Copy-paste the generated prompt

Option C: Standalone Mode (Direct Git Access)

This mode is for AI agents that can directly clone Git repositories and execute shell commands (e.g., Jules by Google, Augment Code Remote Agents). The agent first learns the AGOR protocol:

# For AI agents with direct Git access (e.g., Jules by Google, Augment Code)
# Agent typically executes these commands to understand AGOR protocol:
cd /tmp && git clone https://github.com/jeremiah-k/agor.git && cd agor
cat src/agor/tools/README_ai.md && cat src/agor/tools/AGOR_INSTRUCTIONS.md
# Then, the agent would clone and work on your project repository.

🎭 Step 4: Select Your Role

AGOR will ask you to choose a role. Pick based on what you want to do:

📋 PROJECT COORDINATOR

"I want to plan and coordinate development"

  • Strategic planning and project breakdown
  • Team design and workflow orchestration
  • Multi-agent strategy selection

🔍 Worker Agent

"I want to analyze and work with code"

  • Deep codebase analysis and exploration
  • Implementation and debugging
  • Technical deep-dives

🤖 AGENT WORKER

"I'm following instructions from a coordinator"

  • Task execution and work snapshots
  • Coordination communication
  • Status reporting

🎼 Step 5: Start Coordinating

Once initialized, you'll see a capabilities menu. Try these functions:

# Analyze your codebase
a

# Strategic planning
sp

# Break down project
bp

# Get help anytime
?

🎯 Your First Workflow

Let's walk through a typical coordination session:

Planning Phase (PROJECT COORDINATOR)

1. Type: sp
   → AGOR analyzes your project and creates strategic plan

2. Type: bp
   → Breaks down project into coordinated tasks

3. Type: ss
   → Suggests coordination strategies (Parallel, Pipeline, Swarm, etc.)

4. Type: ct
   → Designs team structure with specialized agents

Implementation Phase (Worker Agent)

1. Type: a
   → Comprehensive codebase analysis and exploration

2. Type: f src/main.py
   → Examine specific files

3. Type: grep "function_name"
   → Search across codebase

4. Type: da
   → Generate detailed snapshot for other agents

💡 Pro Tips

Make It Executable (Google AI Studio)

If you see command simulation instead of execution:

chmod 755 /tmp/agor_tools/git
/tmp/agor_tools/git status

Switch Roles Mid-Session

You can change roles by re-initializing:

# Upload fresh bundle
# Select different role
# Continue with new capabilities

Provide Feedback

Use AGOR's feedback tools to help improve AGOR:

meta
→ Report issues, suggest improvements, share success stories

Memory Synchronization

AGOR automatically manages memory persistence using git branches:

# Memory is automatically synced to dedicated branches
# No manual setup required - works out of the box
# Use standard memory commands:
mem-add    # Add memory entries
mem-search # Search across all memories

🔄 What's Next?

Now that you have AGOR running, explore these advanced topics:

🆘 Need Help?

Something not working?

Want to understand more?

Congratulations! 🎉 You've just set up your first AI coordination workflow. Welcome to the future of development!

Ready to dive deeper? Check out the Multi-Agent Strategies to see what's possible. 🚀