Skip to main content
Install the StackOne AI SDK for Python applications with first-class support for LangChain, CrewAI, and OpenAI tool calling. The package ships typed tool definitions, LangChain/CrewAI adapters, meta tools, dynamic discovery, and LangSmith implicit feedback helpers out of the box.

​
Installation

Install the core SDK: 
# Using uv (Recommended)
uv add stackone-ai

# Using pip
pip install stackone-ai

​
Requirements

  • Python 3.9 or higher for the core SDK
  • Python 3.10+ for dynamic discovery, CrewAI tooling, and implicit feedback helpers
  • uv or pip

​
Optional Extras

Install optional components with extras so you only pull the dependencies you need: 
# Enable dynamic discovery (Python 3.10+)
uv add 'stackone-ai[mcp]'
pip install 'stackone-ai[mcp]'

# Pull LangChain/CrewAI/OpenAI example deps
uv add 'stackone-ai[examples]'
pip install 'stackone-ai[examples]'

# Install everything exposed today
uv add 'stackone-ai[mcp,examples]'
pip install 'stackone-ai[mcp,examples]'

​
Environment Setup

Create a .env file in your project root: 
STACKONE_API_KEY=your_api_key_here
STACKONE_ACCOUNT_ID=optional_default_account

​
Framework Dependencies

Install additional dependencies based on your framework: 
# For LangChain
uv add stackone-ai[examples] langchain langchain-openai
# or using pip
pip install stackone-ai[examples] langchain langchain-openai

# For CrewAI
uv add stackone-ai[examples] crewai
# or using pip
pip install stackone-ai[examples] crewai

# For OpenAI
uv add stackone-ai[examples] openai
# or using pip
pip install stackone-ai[examples] openai

​
Verification

Test your installation: 
from stackone_ai import StackOneToolSet

toolset = StackOneToolSet()
print("✅ StackOne AI SDK installed successfully!")

​
Next Steps