CtrlK
BlogDocsLog inGet started
Tessl Logo

tessl/pypi-langgraph-cli

CLI for interacting with LangGraph API

Overall
score

74%

Evaluation — 74%

↑ 1.17x

Agent success when using this tile

Overview
Eval results
Files

templates.mddocs/

Templates

Project scaffolding system for LangGraph applications. Provides multiple pre-built templates covering common LangGraph patterns and use cases, enabling rapid project initialization with best practices.

Capabilities

Project Creation

Create new LangGraph projects from predefined templates with proper structure and configuration.

def create_new(path: Optional[str], template: Optional[str]) -> None

Purpose: Create a new LangGraph project from a template Parameters:

  • path (Optional[str]): Directory path for the new project (current directory if None)
  • template (Optional[str]): Template identifier (interactive selection if None) Returns: None (creates files and directories) Raises: TemplateError for invalid templates, FileExistsError for existing directories

Usage Examples:

from langgraph_cli.templates import create_new

# Interactive template selection in current directory
create_new(None, None)

# Create project with specific template
create_new("my-agent-app", "react-agent-python")

# Create in custom directory with interactive selection
create_new("./projects/new-bot", None)

Available Templates

New LangGraph Project

Template ID: new-langgraph-project Description: Minimal chatbot with memory Use Case: Basic conversational AI applications with persistent memory

Features:

  • Simple chat interface with conversation memory
  • Basic message handling and response generation
  • Configuration for common LLM providers
  • Development server setup
  • Basic testing structure

Generated Structure:

my-project/
├── langgraph.json          # LangGraph configuration
├── pyproject.toml          # Python project configuration
├── .env.example           # Environment variables template
├── src/
│   ├── __init__.py
│   ├── graph.py           # Main conversation graph
│   └── nodes/             # Graph node implementations
│       ├── __init__.py
│       └── chat.py
├── tests/
│   ├── __init__.py
│   └── test_graph.py
└── README.md

ReAct Agent

Template ID: react-agent-python Description: Extensible agent with tools Use Case: Reasoning and acting agents that can use external tools

Features:

  • ReAct (Reasoning + Acting) pattern implementation
  • Tool integration framework
  • Extensible tool registry
  • Error handling and retry logic
  • Comprehensive logging and observability

Generated Structure:

react-agent/
├── langgraph.json
├── pyproject.toml
├── .env.example
├── src/
│   ├── __init__.py
│   ├── agent.py           # Main ReAct agent
│   ├── tools/             # Tool implementations
│   │   ├── __init__.py
│   │   ├── base.py        # Base tool interface
│   │   ├── web_search.py  # Web search tool
│   │   └── calculator.py  # Math operations
│   └── prompts/
│       ├── __init__.py
│       └── react.py       # ReAct prompting logic
├── tests/
│   ├── __init__.py
│   ├── test_agent.py
│   └── test_tools.py
└── README.md

Memory Agent

Template ID: memory-agent-python Description: ReAct agent with cross-conversation memory Use Case: Agents that remember information across multiple conversations

Features:

  • ReAct pattern with persistent memory
  • Long-term memory storage and retrieval
  • Context management across sessions
  • Memory indexing and search
  • Conversation history management

Generated Structure:

memory-agent/
├── langgraph.json
├── pyproject.toml
├── .env.example
├── src/
│   ├── __init__.py
│   ├── agent.py           # Memory-enabled ReAct agent
│   ├── memory/            # Memory management
│   │   ├── __init__.py
│   │   ├── store.py       # Memory storage interface
│   │   └── retrieval.py   # Memory retrieval logic
│   ├── tools/             # Agent tools
│   │   ├── __init__.py
│   │   └── base.py
│   └── prompts/
│       ├── __init__.py
│       └── memory.py      # Memory-aware prompts
├── tests/
│   ├── __init__.py
│   ├── test_agent.py
│   └── test_memory.py
└── README.md

Retrieval Agent

Template ID: retrieval-agent-python Description: Agent with RAG (Retrieval-Augmented Generation) capabilities Use Case: Agents that can search and reference external knowledge bases

Features:

  • RAG implementation with vector search
  • Document ingestion and indexing
  • Semantic search and retrieval
  • Context-aware response generation
  • Support for multiple document formats

Generated Structure:

retrieval-agent/
├── langgraph.json
├── pyproject.toml
├── .env.example
├── src/
│   ├── __init__.py
│   ├── agent.py           # RAG-enabled agent
│   ├── retrieval/         # RAG implementation
│   │   ├── __init__.py
│   │   ├── embeddings.py  # Embedding generation
│   │   ├── vectorstore.py # Vector storage
│   │   └── retriever.py   # Search and retrieval
│   ├── ingestion/         # Document processing
│   │   ├── __init__.py
│   │   ├── loader.py      # Document loaders
│   │   └── processor.py   # Text processing
│   └── prompts/
│       ├── __init__.py
│       └── rag.py         # RAG-specific prompts
├── data/                  # Sample documents
│   └── sample.txt
├── tests/
│   ├── __init__.py
│   ├── test_agent.py
│   ├── test_retrieval.py
│   └── test_ingestion.py
└── README.md

Data-Enrichment Agent

Template ID: data-enrichment-python Description: Web search and data organization agent Use Case: Agents that gather, process, and organize information from web sources

Features:

  • Web search integration
  • Data extraction and cleaning
  • Information organization and categorization
  • Multi-source data aggregation
  • Structured output generation

Generated Structure:

data-enrichment/
├── langgraph.json
├── pyproject.toml
├── .env.example
├── src/
│   ├── __init__.py
│   ├── agent.py           # Data enrichment agent
│   ├── search/            # Web search capabilities
│   │   ├── __init__.py
│   │   ├── web_search.py  # Search implementation
│   │   └── extractors.py  # Data extraction
│   ├── processing/        # Data processing
│   │   ├── __init__.py
│   │   ├── cleaner.py     # Data cleaning
│   │   └── organizer.py   # Information organization
│   ├── output/            # Output formatting
│   │   ├── __init__.py
│   │   └── formatters.py  # Structured output
│   └── prompts/
│       ├── __init__.py
│       └── enrichment.py  # Data enrichment prompts
├── tests/
│   ├── __init__.py
│   ├── test_agent.py
│   ├── test_search.py
│   └── test_processing.py
└── README.md

Template Selection

Interactive Selection

When no template is specified, the CLI provides an interactive selection interface:

$ langgraph new my-project

? Select a template:
❯ New LangGraph Project - Minimal chatbot with memory
  ReAct Agent - Extensible agent with tools
  Memory Agent - ReAct agent with cross-conversation memory
  Retrieval Agent - Agent with RAG capabilities
  Data-enrichment Agent - Web search and data organization

Direct Template Usage

Templates can be specified directly:

# Create ReAct agent
langgraph new react-bot --template react-agent-python

# Create retrieval agent in specific directory
langgraph new ./agents/rag-bot --template retrieval-agent-python

Template Configuration

The template system uses several constants for management:

# Template definitions with URLs for Python and JavaScript versions
TEMPLATES: dict[str, dict[str, str]] = {
    "New LangGraph Project": {
        "description": "A simple, minimal chatbot with memory.",
        "python": "https://github.com/langchain-ai/new-langgraph-project/archive/refs/heads/main.zip",
        "js": "https://github.com/langchain-ai/new-langgraphjs-project/archive/refs/heads/main.zip",
    },
    "ReAct Agent": {
        "description": "A simple agent that can be flexibly extended to many tools.",
        "python": "https://github.com/langchain-ai/react-agent/archive/refs/heads/main.zip",
        "js": "https://github.com/langchain-ai/react-agent-js/archive/refs/heads/main.zip",
    },
    # Additional templates...
}

# Template identifier mapping for CLI usage
TEMPLATE_ID_TO_CONFIG: dict[str, tuple[str, str, str]]
TEMPLATE_IDS: list[str]

# Help string for CLI template selection
TEMPLATE_HELP_STRING: str

Template Help

The CLI provides detailed template information:

# Template help string with descriptions
TEMPLATE_HELP_STRING = """
Available templates:

  new-langgraph-project     Minimal chatbot with memory
  react-agent-python        Extensible agent with tools
  memory-agent-python       ReAct agent with cross-conversation memory
  retrieval-agent-python    Agent with RAG capabilities
  data-enrichment-python    Web search and data organization

Use 'langgraph new --template TEMPLATE_NAME' to create a project.
"""

Template Components

Each template includes:

Core Files

  • langgraph.json: LangGraph configuration with appropriate dependencies
  • pyproject.toml: Python project configuration with required packages
  • .env.example: Environment variables template with provider APIs
  • README.md: Template-specific setup and usage instructions

Source Code Structure

  • Main agent/graph file: Core application logic
  • Node implementations: Modular graph node functions
  • Tool implementations: External tool integrations
  • Prompt templates: LLM prompt engineering
  • Configuration modules: Settings and configuration management

Testing Framework

  • Unit tests: Component-level testing
  • Integration tests: End-to-end workflow testing
  • Test fixtures: Sample data and mock objects
  • Test configuration: Testing-specific settings

Documentation

  • README: Quick start and overview
  • Code comments: Inline documentation
  • Configuration examples: Sample configurations
  • Deployment guides: Production deployment instructions

Customization

Templates serve as starting points and can be customized:

Configuration Customization

Modify langgraph.json for:

  • Different Python versions
  • Additional dependencies
  • Custom environment variables
  • Production deployment settings

Code Customization

Templates provide extensible architectures:

  • Add new tools to ReAct agents
  • Implement custom memory backends
  • Add document loaders to retrieval agents
  • Extend data processing pipelines

Deployment Customization

Each template includes deployment options:

  • Docker configuration
  • Environment-specific settings
  • CI/CD pipeline examples
  • Monitoring and logging setup

Best Practices

Templates follow LangGraph best practices:

Code Organization

  • Clear separation of concerns
  • Modular node implementations
  • Reusable components
  • Consistent naming conventions

Configuration Management

  • Environment-based configuration
  • Secure secret handling
  • Flexible deployment options
  • Development/production separation

Testing Strategy

  • Comprehensive test coverage
  • Mock external dependencies
  • Integration test patterns
  • Performance testing examples

Error Handling

  • Graceful failure handling
  • Informative error messages
  • Recovery mechanisms
  • Logging and observability

Install with Tessl CLI

npx tessl i tessl/pypi-langgraph-cli

docs

cli-commands.md

configuration.md

docker-integration.md

index.md

templates.md

tile.json