Contributing to LangGraph

Thank you for your interest in contributing to LangGraph! This guide will help you get started.

Development Setup

Prerequisites

• Python 3.9+
• Poetry for dependency management

Installation

git clone https://github.com/langchain-ai/langgraph.git
cd langgraph
poetry install --with dev,test

Running Tests

# Run all tests
poetry run pytest

# Run specific test file
poetry run pytest tests/unit_tests/test_graph.py

# Run with coverage
poetry run pytest --cov=langgraph

Contribution Workflow

1. Check existing issues: Look for open issues or discussions before starting work
2. Create an issue: For significant changes, open an issue to discuss the approach
3. Fork and branch: Create a feature branch from main
4. Implement: Write your code following our style guidelines
5. Test: Add tests for new functionality, ensure all existing tests pass
6. Document: Update docstrings and documentation as needed
7. Submit PR: Open a pull request with a clear description

Pull Request Standards

Required

• All CI checks must pass (lint, type check, tests)
• New features must include test coverage
• Public API changes must include docstring updates
• Breaking changes must be discussed in an issue first

PR Description Template

• What: Brief description of the change
• Why: Motivation and context
• How: Implementation approach
• Testing: How the change was tested

Code Quality

• Format with black and ruff
• Type check with mypy --strict
• Sort imports with isort
• Lint with ruff check

Areas for Contribution

• New checkpoint backends (Redis, DynamoDB, etc.)
• Example agent implementations
• Documentation improvements and tutorials
• Performance optimizations in the Pregel execution engine
• Bug fixes and test coverage improvements

License

By contributing, you agree that your contributions will be licensed under the MIT License.