openstackai 0.6.1

The Pandas of AI Agents - Build AI applications in 3 lines or less

🧠 openstackai

Three-Dimensional Intelligence Engine

The Intelligence Engine for Software Factories
Build, Orchestrate, and Scale AI-Native Applications

---

🚀 Quick Start

pip install openstackai

One-Line AI Operations

from openstackai import ask, summarize, research

# Ask anything
answer = ask("What is machine learning?")

# Summarize documents
summary = summarize("Long document text...")

# Research topics
report = research("Latest AI trends 2026")

Build AI Agents

from openstackai import Agent, Runner

agent = Agent(
    name="Assistant",
    instructions="You are a helpful AI assistant."
)

result = Runner.run_sync(agent, "Hello!")
print(result.final_output)

Multi-Agent Workflows

from openstackai import Agent, Runner
from openstackai.blueprint import ChainWorkflow

researcher = Agent(name="Researcher", instructions="Research topics")
writer = Agent(name="Writer", instructions="Write articles")

workflow = ChainWorkflow([researcher, writer])
result = Runner.run_sync(workflow, "AI in healthcare")

---

✨ Key Features

Feature	Description
🎯 Simple APIs	One-liner functions: ask(), summarize(), research(), extract(), generate()
🤖 Agent Framework	Full agent system with Agent, Runner, memory, and context passing
🔗 Multi-Agent	Orchestration patterns: Chain, Supervisor, Swarm, Tree of Thoughts
🔌 Multi-Provider	OpenAI, Azure OpenAI, Anthropic, Ollama, Gemini, Groq
📊 Vector DBs	ChromaDB, Pinecone, Qdrant, Weaviate integration
🏢 Enterprise	Azure AD auth, sessions, guardrails, evaluation, tracing
🧩 Extensible	Custom skills, OpenAPI tools, A2A protocol, plugins
🎙️ Multimodal	Vision, audio, document understanding

---

🔌 Supported Providers

Provider	Status	Authentication
OpenAI	✅ Full Support	API Key
Azure OpenAI	✅ Full Support	API Key / Azure AD
Anthropic Claude	✅ Full Support	API Key
Ollama	✅ Full Support	Local / Remote
Google Gemini	✅ Full Support	API Key
Groq	✅ Full Support	API Key

---

📦 Module Overview

Module	Purpose
openstackai.easy	One-liner AI functions
openstackai.core	Agent, LLM providers, memory
openstackai.runner	Agent execution engine
openstackai.blueprint	Workflows & orchestration patterns
openstackai.skills	Tools and skills system
openstackai.sessions	SQLite/Redis session storage
openstackai.vectordb	Vector database connectors
openstackai.evaluation	Agent evaluation framework
openstackai.openapi	OpenAPI tool generation
openstackai.multimodal	Vision, audio support
openstackai.a2a	Agent-to-Agent protocol

---

🔧 Configuration

Environment Variables

# OpenAI
export OPENAI_API_KEY="sk-..."

# Azure OpenAI
export AZURE_OPENAI_ENDPOINT="https://your-resource.openai.azure.com/"
export AZURE_OPENAI_DEPLOYMENT="gpt-4o-mini"
export AZURE_OPENAI_API_KEY="..."  # Or use Azure AD

# Anthropic
export ANTHROPIC_API_KEY="sk-ant-..."

YAML Agent Configuration

# research_agent.yaml
name: "Research Agent"
instructions: "You are an expert researcher."
provider: "openai"
model: "gpt-4o-mini"
skills:
  - web_search
  - summarize

from openstackai.config import AgentConfig
agent = AgentConfig.from_yaml("research_agent.yaml").create_agent()

---

📚 Documentation

• 📖 Full Documentation: gitpavleenbali.github.io/PYAI
• 📘 GitHub Repository: github.com/gitpavleenbali/PYAI
• 📋 Changelog: Changelog
• 🏗️ Architecture: Architecture Guide

---

🤝 Contributing

Contributions are welcome! See our Contributing Guide.

---

📄 License

MIT License - see LICENSE for details.

---

Built with ❤️ for the AI community
openstackai - The Intelligence Engine for Software Factories