# Install
pip install arklex

# Create .env file
echo "OPENAI_API_KEY=your_key_here" > .env

# Clone the repository (required if you want to run example/test scripts)
git clone https://github.com/arklexai/Agent-First-Organization.git
cd Agent-First-Organization
pip install -e .

# Test your API keys (recommended)
python test_api_keys.py

# Create your first agent
python create.py \
  --config ./examples/customer_service/customer_service_config.json \
  --output-dir ./examples/customer_service \
  --llm_provider openai \
  --model gpt-4o

# Run agent
python run.py \
  --input-dir ./examples/customer_service \
  --llm_provider openai \
  --model gpt-4o

# For evaluation and testing, you can also use the model API server:
# 1. Start the model API server (defaults to OpenAI with "gpt-4o-mini" model):
python model_api.py --input-dir ./examples/customer_service

# 2. Run evaluation (in a separate terminal):
python eval.py --model_api http://127.0.0.1:8000/eval/chat \
  --config "examples/customer_service/customer_service_config.json" \
  --documents_dir "examples/customer_service" \
  --model "claude-3-haiku-20240307" \
  --llm_provider "anthropic" \
  --task "all"

👉 Explore the full tutorial: Customer Service Agent Walkthrough

• 🚀 90% Faster Development — Deploy agents in days, not months
• 🧠 Agent-First Design — Purpose-built for multi-agent orchestration
• 🔌 Model Agnostic — OpenAI, Anthropic, Gemini, and more
• 📊 Built-in Evaluation — Comprehensive testing suite
• 🛡️ Enterprise Security — Authentication and rate limiting
• ⚡ Production Ready — Monitoring, logging, auto-scaling

graph TB
    A[Task Graph] --> B[Orchestrator]
    B --> C[Workers]
    B --> D[Tools]
    C --> E[RAG Worker]
    C --> F[Database Worker]
    C --> G[Custom Workers]
    D --> I[API Tools]
    D --> J[External Tools]

Core Components:

• Task Graph — Declarative DAG workflows
• Orchestrator — Runtime engine with state management
• Workers — RAG, database, web automation
• Tools — Shopify, HubSpot, Google Calendar integrations

Requirements: Python 3.10+, API keys

# Required: Choose one or more LLM providers
OPENAI_API_KEY=your_key_here
# OR ANTHROPIC_API_KEY=your_key_here
# OR GOOGLE_API_KEY=your_key_here

# Optional: Enhanced features
MILVUS_URI=your_milvus_uri
MYSQL_USERNAME=your_username
TAVILY_API_KEY=your_tavily_key

Testing API Keys: After adding your API keys to the .env file, run the test script to verify they work correctly:

# Test all configured API keys
python test_api_keys.py

# Test specific providers only
python test_api_keys.py --providers openai gemini
python test_api_keys.py --providers openai anthropic

• 📚 Full Documentation
• 🚀 Quick Start
• 🛠️ API Reference
• 🏗️ Architecture
• 🚀 Deployment

• 🐛 Report Issues
• 💬 Discord
• 🐦 Twitter
• 💼 LinkedIn
• 📧 Email Support

Arklex AI is released under the MIT License. See LICENSE for details.

This means you can:

• ✅ Use Arklex AI for commercial projects
• ✅ Modify and distribute the code
• ✅ Use it in proprietary applications
• ✅ Sell applications built with Arklex AI

The only requirement is that you include the original license and copyright notice.

Thanks to all our contributors and the open-source community for making this project possible!

Arklex AI builds on the shoulders of giants:

• LangChain — LLM framework and tooling
• FastAPI — Modern web framework
• Pydantic — Data validation
• SQLAlchemy — Database ORM
• Milvus — Vector database
• And many more...