LangGraph Starter

A LangGraph-based starter template for building collections of AI agents. This template provides a foundation for creating multi-agent systems with capabilities like LinkedIn lead collection, web search, and Airtable data integration.

---

✨ Features

• Collection of configurable AI agents using DeepAgents and LangGraph
• Agent entrypoints defined in langgraph.json for easy management
• Example: LinkedIn Lead Collector agent
• Airtable integration for structured data storage
• Built-in web search tools
• VS Code debugging configuration included
• Docker-ready for containerized development & deployment

---

🚀 Quick Start

Prerequisites

• Python 3.11+
• Docker (optional, for containerized deployment)

Installation

1. Clone the repository

   git clone <YOUR_REPO_URL> langgraph-starter
   cd langgraph-starter

2. Install dependencies with uv (recommended)

   uv venv
   source .venv/bin/activate
   uv sync

3. Set up environment variables

   cp .env.example .env
   # Edit .env with your API keys and configuration

4. Start the development server

   make dev

MCP Server

This project includes an MCP (Model Context Protocol) server for enhanced AI integrations.

Start the MCP server:

python mcp

---

🖥️ Recommended LangGraph UIs

• LangGraph Studio — Official Studio for visual graph editing & monitoring
• Agent Chat UI — Clean chat interface for testing your agents

---

🛠️ Development & Debugging

• .vscode/launch.json is pre-configured for:

  • Attach to LangGraph — Debug a running server on port 5678
  • Debug Script — Run and debug individual scripts

Steps:

1. Run:

   make debug

2. In VS Code, press F5 and select Attach to LangGraph.

---

🗂️ Project Structure

langgraph-starter/
├─ agents/
│  └─ linkedin_leads.py          # LinkedIn lead collection agent (example)
├─ tools/
│  ├─ __init__.py
│  ├─ airtable.py                # Airtable integration tools
│  └─ search.py                  # Web search tools
├─ .vscode/
│  └─ launch.json                # VS Code debug config
├─ docker-compose.yml            # Docker services
├─ langgraph.json                # LangGraph configuration
├─ Makefile                      # Dev commands
├─ pyproject.toml                # Python project configuration
└─ .env.example                  # Example environment variables

---

🧰 Commands

Make targets

make dev      # Start development server
make debug    # Start with debugging enabled (port 5678)
make build    # Build and push Docker image

Using the LangGraph CLI directly

langgraph dev                       # Start dev server
langgraph dev --debug-port 5678     # Start with debugging
langgraph build                     # Build the application

---

🧠 How It Works

Agent Architecture

• Modular Design — Each agent is a separate module under agents/
• LangGraph Integration — Agents and entrypoints are declared in langgraph.json
• Tooling — Shared tools under tools/ (e.g., Airtable, search) can be reused
• State & Orchestration — LangGraph coordinates agent state and message flow

Included Example Agents

1. LinkedIn Lead Collector — Searches LinkedIn for potential leads (example logic)
2. Note Taker — Normalizes and organizes collected data
3. Airtable Integration — Persists structured results to Airtable

Adding a New Agent

1. Create a new file under agents/ (e.g., my_agent.py)
2. Register it in langgraph.json (entrypoints, graph configuration)
3. Wire up any shared tools you need from tools/

---

⚙️ Configuration

Key files:

• langgraph.json — Graphs and entrypoints
• .env — Environment variables and API keys
• pyproject.toml — Python dependencies and project metadata
• .vscode/launch.json — VS Code debugging configuration

Tip: Ensure any secrets (API keys, tokens) are only in .env and excluded from version control.

---

🐳 Docker (optional)

Build and run with Docker:

docker compose up --build

Mount local files (e.g., /files) via docker-compose.yml volumes for read/write access.

---

🤝 Contributing

1. Fork the repo

2. Create a feature branch:

   git checkout -b feature/amazing-feature

3. Commit:

   git commit -m "Add amazing feature"

4. Push:

   git push origin feature/amazing-feature

5. Open a Pull Request

---

📄 License

This project is licensed under the terms specified in the LICENSE file.