### Start Development Environment (Bash)

Source: https://github.com/francescofano/langgraph-telegram-bot/blob/main/README.md

Command to start the development environment with hot-reloading capabilities using Docker Compose.

```bash
docker-compose -f docker-compose.dev.yml up --build
```

--------------------------------

### Clone and Run Telegram Bot (Bash)

Source: https://github.com/francescofano/langgraph-telegram-bot/blob/main/README.md

Commands to clone the repository, set up the environment by copying and editing the .env file, and start the bot's services using Docker Compose.

```bash
git clone https://github.com/francescofano/langgraph-telegram-bot.git
cd langgraph-telegram-bot
cp .env.example .env
# Edit .env with your credentials
docker-compose up --build
```

--------------------------------

### Start Telegram Bot Services (Bash)

Source: https://github.com/francescofano/langgraph-telegram-bot/blob/main/README.md

Command to start the bot's services using a specific Docker Compose file.

```bash
docker-compose -f docker-compose.yml up --build
```

--------------------------------

### Implement Advanced LangGraph (Python)

Source: https://github.com/francescofano/langgraph-telegram-bot/blob/main/README.md

Python code showing how to set up an advanced LangGraph by defining nodes, conditional edges, and the entry point within the AgentFactory.

```python
@staticmethod
async def create_advanced_graph(...):
    graph = StateGraph(Any)
    
    
    graph.add_node("classify_intent", classify_user_intent)
    graph.add_node("answer_question", answer_general_question)
    
    # Define edges
    graph.add_conditional_edges(
        "classify_intent",
        route_by_intent,
        {
            "question": "answer_question",
            "search": "search_knowledge"
        }
    )
    
    # Set the entry point
    graph.set_entry_point("classify_intent")
    
    return graph
```

--------------------------------

### Environment Variables for Telegram Bot (Bash)

Source: https://github.com/francescofano/langgraph-telegram-bot/blob/main/README.md

Lists essential environment variables for configuring the Telegram bot, including database connection strings, Telegram token, OpenAI API key, and model configurations.

```bash
# Database configuration
PG_CONNECTION_STRING=postgresql://user:pass@host:port/db
REDIS_URL=redis://redis:6379/0

# Telegram configuration
TELEGRAM_TOKEN=your_telegram_bot_token
DEBOUNCE_TIME=5.0
LLM_CALLS_PER_MINUTE=5

# Agent configuration
LLM_MODEL=gpt-4o-mini
EMBED_MODEL=openai:text-embedding-3-small
VECTOR_DIMS=1536

# API Keys
OPENAI_API_KEY=your_openai_key
```

--------------------------------

### Scale Bot Instances and Add Healthchecks (Bash)

Source: https://github.com/francescofano/langgraph-telegram-bot/blob/main/README.md

Bash commands for scaling the bot instances using Docker Compose and checking the health of the running services.

```bash
# Scale bot instances
docker-compose up --scale bot=3 -d

# Add healthchecks
curl http://localhost:8000/health
```

--------------------------------

### Add Custom Tool to Agent (Python)

Source: https://github.com/francescofano/langgraph-telegram-bot/blob/main/README.md

Python code snippet demonstrating how to add a custom tool to the agent's capabilities by modifying the AgentFactory class.

```python
# Example: Adding a new tool to the agent
from langmem import create_manage_memory_tool
from some_package import create_custom_tool

class AgentFactory:
    @staticmethod
    async def create_agent(...):
        return create_react_agent(
            f"openai:{llm_model}",
            prompt=AgentFactory.create_prompt,
            tools=[
                create_manage_memory_tool(namespace=("memories",)),
                create_custom_tool()  # Add your custom tool here
            ],
            checkpointer=checkpointer,
            store=store,
        )
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.