Skip to main content

Contributing to LangChainGo

Thank you for your interest in contributing to LangChainGo! This guide helps you get started.

Ways to contribute​

1. Code contributions​

  • Bug fixes: Help us squash bugs and improve stability
  • New features: Implement new LLM providers, tools, or chains
  • Performance improvements: Optimize existing code
  • Tests: Improve test coverage and add missing tests

2. Documentation contributions​

  • Tutorials: Write step-by-step guides for common use cases
  • How-to guides: Create practical solutions for specific problems
  • API documentation: Improve code comments and examples
  • Conceptual guides: Explain architectural decisions and patterns

3. Community support​

  • Answer questions: Help others in GitHub Discussions
  • Report issues: File detailed bug reports
  • Review PRs: Provide feedback on pull requests
  • Share examples: Showcase your LangChainGo projects

Getting started​

Development setup​

  1. Fork the repository on GitHub

  2. Clone your fork locally:

    git clone https://github.com/YOUR-USERNAME/langchaingo.git
    cd langchaingo

  3. Add the upstream remote:

    git remote add upstream https://github.com/tmc/langchaingo.git

  4. Create a feature branch:

    git checkout -b feature/your-feature-name

Code style​

  • Follow standard Go conventions and idioms
  • Run go fmt before committing
  • Ensure all tests pass with go test ./...
  • Add tests for new functionality
  • Use package-prefixed commit messages (see PR Guidelines below)
  • Keep commits focused to a single topic

Testing​

When contributing code that interacts with external APIs:

  1. Use the internal httprr tool for recording HTTP interactions
  2. Never commit real API keys or secrets
  3. Ensure tests can run without external dependencies
  4. See the Architecture Guide for details

Contribution process​

  1. Check existing issues: Look for existing issues or discussions about your idea
  2. Open an issue: For significant changes, open an issue to discuss first
  3. Make changes: Implement your changes in a feature branch
  4. Follow commit style: Use Go-style package-prefixed commit messages
  5. Test thoroughly: Ensure all tests pass and add new ones as needed
  6. Submit PR: Open a pull request with a clear description following our guidelines
  7. Address feedback: Respond to review comments promptly

Pull request guidelines​

PR title format​

Use Go-style package-prefixed commit messages following the Go Contribute Guidelines:

  • memory: add interfaces for custom storage backends
  • llms/openai: fix streaming response handling
  • chains: implement conversation chain with memory
  • vectorstores/chroma: add support for metadata filtering
  • docs: update getting started guide for new API
  • agents: add tool calling support for GPT-4
  • examples: add RAG implementation tutorial

Format: package: description in lowercase without period

Examples of good commit messages:

  • llms/anthropic: implement function calling support
  • memory: fix buffer overflow in conversation memory
  • tools: add calculator tool with error handling
  • all: update dependencies and organize go.mod file

PR description​

Include:

  • Summary of changes
  • Related issue numbers
  • Testing performed
  • Breaking changes (if any)
  • Reference to similar features in Python/TypeScript LangChain (when applicable)

Documentation contributions​

See our dedicated Documentation Contribution Guide for details on:

  • Writing tutorials
  • Creating how-to guides
  • Documentation style guide
  • Building and testing docs locally

Code of conduct​

Please note that this project follows a code of conduct. By participating, you are expected to uphold this code. Please report unacceptable behavior to the project maintainers.

Recognition​

Contributors are recognized in:

  • The project's contributor list
  • Release notes for significant contributions
  • Documentation credits for written content

Questions?​

Thank you for helping make LangChainGo better!