LangChain Learning Repository

This repository contains code examples and implementations following the LangChain JS/TS tutorials.

What is LangChain?

LangChain is a framework for developing applications powered by language models. It enables the creation of applications that are:

• Context-aware: Connect language models to sources of data
• Reasoning: Use language models to make decisions and take actions

Purpose

This repository serves as a learning playground for exploring LangChain's capabilities. It includes implementations of various tutorials and examples from the official documentation, allowing for hands-on experience with:

• Building conversational agents
• Creating chains and sequences
• Implementing retrieval-augmented generation (RAG)
• Working with various language models
• Utilizing vector stores and embedding techniques

Getting Started

Prerequisites

• Bun v1.2.4 or later

Installation

To install dependencies:

bun install

Running Examples

To run the default example:

bun run index.ts

For specific examples:

bun run examples/<example-name>.ts

Project Structure

• examples/ - Individual tutorial implementations
• src/ - Shared components and utilities
• data/ - Sample data used by examples

Tutorial Implementations

This repository follows tutorials from the LangChain documentation, including:

• Chat models and chat prompts
• Building conversational agents
• Retrieval-augmented generation
• Working with structured outputs
• Memory and state management

Featured Projects

Crypto Contract Analysis Supervisor

The repository includes an advanced implementation of a multi-agent system for comprehensive crypto contract analysis:

• Architecture: Uses LangGraph's supervisor pattern to coordinate specialized agents

• Workflow:

  1. Data collection from multiple sources (social, market, on-chain)
  2. Multi-faceted analysis through specialized agents
  3. Opportunity evaluation based on combined insights
  4. Detailed report generation

• Key Components:

  • Data Fetcher Agent: Retrieves data from Farcaster, Coingecko, and Etherscan
  • Analysis Agents: Separate agents for social, market, and on-chain analysis
  • Evaluation Agent: Assesses investment opportunity based on all analyses
  • Report Generation Agent: Creates comprehensive markdown reports

To run the contract analysis:

bun run tests/onchain-supervisor/run_contract_analysis.ts

Resources

• LangChain Documentation
• LangChain GitHub Repository
• LangChain Discord Community

License

This project is intended for learning purposes.

---

This project was created using bun init in bun v1.2.4. Bun is a fast all-in-one JavaScript runtime.