langchain 0.4.0.dev0__tar.gz → 1.0.0a2__tar.gz

{langchain-0.4.0.dev0 → langchain-1.0.0a2}/PKG-INFO

@@ -1,26 +1,24 @@
 Metadata-Version: 2.1
 Name: langchain
-Version: 0.4.0.dev0
+Version: 1.0.0a2
 Summary: Building applications with LLMs through composability
 License: MIT
 Project-URL: Source Code, https://github.com/langchain-ai/langchain/tree/master/libs/langchain
 Project-URL: Release Notes, https://github.com/langchain-ai/langchain/releases?q=tag%3A%22langchain%3D%3D0%22&expanded=true
 Project-URL: repository, https://github.com/langchain-ai/langchain
-Requires-Python: <4.0,>=3.9
-Requires-Dist: langchain-core<1.0.0,>=0.4.0.dev0
+Requires-Python: >=3.10
+Requires-Dist: langchain-core<2.0.0,>=0.3.75
 Requires-Dist: langchain-text-splitters<1.0.0,>=0.3.9
-Requires-Dist: langsmith>=0.1.17
-Requires-Dist: pydantic<3.0.0,>=2.7.4
-Requires-Dist: SQLAlchemy<3,>=1.4
-Requires-Dist: requests<3,>=2
-Requires-Dist: PyYAML>=5.3
-Requires-Dist: async-timeout<5.0.0,>=4.0.0; python_version < "3.11"
-Provides-Extra: community
-Requires-Dist: langchain-community; extra == "community"
+Requires-Dist: langgraph>=0.6.0
+Requires-Dist: pydantic>=2.7.4
 Provides-Extra: anthropic
 Requires-Dist: langchain-anthropic; extra == "anthropic"
 Provides-Extra: openai
 Requires-Dist: langchain-openai; extra == "openai"
+Provides-Extra: azure-ai
+Requires-Dist: langchain-azure-ai; extra == "azure-ai"
+Provides-Extra: google-vertexai
+Requires-Dist: langchain-google-vertexai; extra == "google-vertexai"
 Provides-Extra: google-genai
 Requires-Dist: langchain-google-genai; extra == "google-genai"
 Provides-Extra: fireworks
@@ -35,8 +33,12 @@ Provides-Extra: huggingface
 Requires-Dist: langchain-huggingface; extra == "huggingface"
 Provides-Extra: groq
 Requires-Dist: langchain-groq; extra == "groq"
+Provides-Extra: aws
+Requires-Dist: langchain-aws; extra == "aws"
 Provides-Extra: deepseek
 Requires-Dist: langchain-deepseek; extra == "deepseek"
+Provides-Extra: xai
+Requires-Dist: langchain-xai; extra == "xai"
 Provides-Extra: perplexity
 Requires-Dist: langchain-perplexity; extra == "perplexity"
 Description-Content-Type: text/markdown
@@ -46,29 +48,21 @@ Description-Content-Type: text/markdown
 ⚡ Building applications with LLMs through composability ⚡
 
 [![Release Notes](https://img.shields.io/github/release/langchain-ai/langchain)](https://github.com/langchain-ai/langchain/releases)
-[![lint](https://github.com/langchain-ai/langchain/actions/workflows/lint.yml/badge.svg)](https://github.com/langchain-ai/langchain/actions/workflows/lint.yml)
-[![test](https://github.com/langchain-ai/langchain/actions/workflows/test.yml/badge.svg)](https://github.com/langchain-ai/langchain/actions/workflows/test.yml)
 [![Downloads](https://static.pepy.tech/badge/langchain/month)](https://pepy.tech/project/langchain)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 [![Twitter](https://img.shields.io/twitter/url/https/twitter.com/langchainai.svg?style=social&label=Follow%20%40LangChainAI)](https://twitter.com/langchainai)
 [![Open in Dev Containers](https://img.shields.io/static/v1?label=Dev%20Containers&message=Open&color=blue&logo=visualstudiocode)](https://vscode.dev/redirect?url=vscode://ms-vscode-remote.remote-containers/cloneInVolume?url=https://github.com/langchain-ai/langchain)
 [![Open in GitHub Codespaces](https://github.com/codespaces/badge.svg)](https://codespaces.new/langchain-ai/langchain)
 [![GitHub star chart](https://img.shields.io/github/stars/langchain-ai/langchain?style=social)](https://star-history.com/#langchain-ai/langchain)
-[![Dependency Status](https://img.shields.io/librariesio/github/langchain-ai/langchain)](https://libraries.io/github/langchain-ai/langchain)
-[![Open Issues](https://img.shields.io/github/issues-raw/langchain-ai/langchain)](https://github.com/langchain-ai/langchain/issues)
-
 
 Looking for the JS/TS version? Check out [LangChain.js](https://github.com/langchain-ai/langchainjs).
 
-To help you ship LangChain apps to production faster, check out [LangSmith](https://smith.langchain.com). 
-[LangSmith](https://smith.langchain.com) is a unified developer platform for building, testing, and monitoring LLM applications. 
-Fill out [this form](https://www.langchain.com/contact-sales) to speak with our sales team.
+To help you ship LangChain apps to production faster, check out [LangSmith](https://smith.langchain.com).
+[LangSmith](https://smith.langchain.com) is a unified developer platform for building, testing, and monitoring LLM applications.
 
 ## Quick Install
 
 `pip install langchain`
-or
-`pip install langsmith && conda install langchain -c conda-forge`
 
 ## 🤔 What is this?
 
@@ -78,22 +72,22 @@ This library aims to assist in the development of those types of applications. C
 
 **❓ Question answering with RAG**
 
-- [Documentation](https://python.langchain.com/docs/use_cases/question_answering/)
+- [Documentation](https://python.langchain.com/docs/tutorials/rag/)
 - End-to-end Example: [Chat LangChain](https://chat.langchain.com) and [repo](https://github.com/langchain-ai/chat-langchain)
 
 **🧱 Extracting structured output**
 
-- [Documentation](https://python.langchain.com/docs/use_cases/extraction/)
+- [Documentation](https://python.langchain.com/docs/tutorials/extraction/)
 - End-to-end Example: [SQL Llama2 Template](https://github.com/langchain-ai/langchain-extract/)
 
 **🤖 Chatbots**
 
-- [Documentation](https://python.langchain.com/docs/use_cases/chatbots)
+- [Documentation](https://python.langchain.com/docs/tutorials/chatbot/)
 - End-to-end Example: [Web LangChain (web researcher chatbot)](https://weblangchain.vercel.app) and [repo](https://github.com/langchain-ai/weblangchain)
 
 ## 📖 Documentation
 
-Please see [here](https://python.langchain.com) for full documentation on:
+Please see [our full documentation](https://python.langchain.com) on:
 
 - Getting started (installation, setting up the environment, simple examples)
 - How-To examples (demos, integrations, helper functions)
@@ -123,7 +117,7 @@ Agents involve an LLM making decisions about which Actions to take, taking that
 
 **🧐 Evaluation:**
 
-[BETA] Generative models are notoriously hard to evaluate with traditional metrics. One new way of evaluating them is using language models themselves to do the evaluation. LangChain provides some prompts/chains for assisting in this.
+Generative models are notoriously hard to evaluate with traditional metrics. One new way of evaluating them is using language models themselves to do the evaluation. LangChain provides some prompts/chains for assisting in this.
 
 For more information on these concepts, please see our [full documentation](https://python.langchain.com).
 

{langchain-0.4.0.dev0 → langchain-1.0.0a2}/README.md

@@ -3,29 +3,21 @@
 ⚡ Building applications with LLMs through composability ⚡
 
 [![Release Notes](https://img.shields.io/github/release/langchain-ai/langchain)](https://github.com/langchain-ai/langchain/releases)
-[![lint](https://github.com/langchain-ai/langchain/actions/workflows/lint.yml/badge.svg)](https://github.com/langchain-ai/langchain/actions/workflows/lint.yml)
-[![test](https://github.com/langchain-ai/langchain/actions/workflows/test.yml/badge.svg)](https://github.com/langchain-ai/langchain/actions/workflows/test.yml)
 [![Downloads](https://static.pepy.tech/badge/langchain/month)](https://pepy.tech/project/langchain)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 [![Twitter](https://img.shields.io/twitter/url/https/twitter.com/langchainai.svg?style=social&label=Follow%20%40LangChainAI)](https://twitter.com/langchainai)
 [![Open in Dev Containers](https://img.shields.io/static/v1?label=Dev%20Containers&message=Open&color=blue&logo=visualstudiocode)](https://vscode.dev/redirect?url=vscode://ms-vscode-remote.remote-containers/cloneInVolume?url=https://github.com/langchain-ai/langchain)
 [![Open in GitHub Codespaces](https://github.com/codespaces/badge.svg)](https://codespaces.new/langchain-ai/langchain)
 [![GitHub star chart](https://img.shields.io/github/stars/langchain-ai/langchain?style=social)](https://star-history.com/#langchain-ai/langchain)
-[![Dependency Status](https://img.shields.io/librariesio/github/langchain-ai/langchain)](https://libraries.io/github/langchain-ai/langchain)
-[![Open Issues](https://img.shields.io/github/issues-raw/langchain-ai/langchain)](https://github.com/langchain-ai/langchain/issues)
-
 
 Looking for the JS/TS version? Check out [LangChain.js](https://github.com/langchain-ai/langchainjs).
 
-To help you ship LangChain apps to production faster, check out [LangSmith](https://smith.langchain.com). 
-[LangSmith](https://smith.langchain.com) is a unified developer platform for building, testing, and monitoring LLM applications. 
-Fill out [this form](https://www.langchain.com/contact-sales) to speak with our sales team.
+To help you ship LangChain apps to production faster, check out [LangSmith](https://smith.langchain.com).
+[LangSmith](https://smith.langchain.com) is a unified developer platform for building, testing, and monitoring LLM applications.
 
 ## Quick Install
 
 `pip install langchain`
-or
-`pip install langsmith && conda install langchain -c conda-forge`
 
 ## 🤔 What is this?
 
@@ -35,22 +27,22 @@ This library aims to assist in the development of those types of applications. C
 
 **❓ Question answering with RAG**
 
-- [Documentation](https://python.langchain.com/docs/use_cases/question_answering/)
+- [Documentation](https://python.langchain.com/docs/tutorials/rag/)
 - End-to-end Example: [Chat LangChain](https://chat.langchain.com) and [repo](https://github.com/langchain-ai/chat-langchain)
 
 **🧱 Extracting structured output**
 
-- [Documentation](https://python.langchain.com/docs/use_cases/extraction/)
+- [Documentation](https://python.langchain.com/docs/tutorials/extraction/)
 - End-to-end Example: [SQL Llama2 Template](https://github.com/langchain-ai/langchain-extract/)
 
 **🤖 Chatbots**
 
-- [Documentation](https://python.langchain.com/docs/use_cases/chatbots)
+- [Documentation](https://python.langchain.com/docs/tutorials/chatbot/)
 - End-to-end Example: [Web LangChain (web researcher chatbot)](https://weblangchain.vercel.app) and [repo](https://github.com/langchain-ai/weblangchain)
 
 ## 📖 Documentation
 
-Please see [here](https://python.langchain.com) for full documentation on:
+Please see [our full documentation](https://python.langchain.com) on:
 
 - Getting started (installation, setting up the environment, simple examples)
 - How-To examples (demos, integrations, helper functions)
@@ -80,7 +72,7 @@ Agents involve an LLM making decisions about which Actions to take, taking that
 
 **🧐 Evaluation:**
 
-[BETA] Generative models are notoriously hard to evaluate with traditional metrics. One new way of evaluating them is using language models themselves to do the evaluation. LangChain provides some prompts/chains for assisting in this.
+Generative models are notoriously hard to evaluate with traditional metrics. One new way of evaluating them is using language models themselves to do the evaluation. LangChain provides some prompts/chains for assisting in this.
 
 For more information on these concepts, please see our [full documentation](https://python.langchain.com).
 

langchain-1.0.0a2/langchain/__init__.py

@@ -0,0 +1,26 @@
+"""Main entrypoint into LangChain."""
+
+from typing import Any
+
+__version__ = "1.0.0a1"
+
+
+def __getattr__(name: str) -> Any:  # noqa: ANN401
+    """Get an attribute from the package.
+
+    TODO: will be removed in a future alpha version.
+    """
+    if name == "verbose":
+        from langchain.globals import _verbose
+
+        return _verbose
+    if name == "debug":
+        from langchain.globals import _debug
+
+        return _debug
+    if name == "llm_cache":
+        from langchain.globals import _llm_cache
+
+        return _llm_cache
+    msg = f"Could not find: {name}"
+    raise AttributeError(msg)

langchain-1.0.0a2/langchain/_internal/_documents.py

@@ -0,0 +1,35 @@
+"""Internal document utilities."""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from langchain_core.documents import Document
+
+
+def format_document_xml(doc: Document) -> str:
+    """Format a document as XML-like structure for LLM consumption.
+
+    Args:
+        doc: Document to format
+
+    Returns:
+        Document wrapped in XML tags:
+            <document>
+                <id>...</id>
+                <content>...</content>
+                <metadata>...</metadata>
+            </document>
+
+    .. note::
+        Does not generate valid XML or escape special characters. Intended for
+        semi-structured LLM input only.
+
+    """
+    id_str = f"<id>{doc.id}</id>" if doc.id is not None else "<id></id>"
+    metadata_str = ""
+    if doc.metadata:
+        metadata_items = [f"{k}: {v!s}" for k, v in doc.metadata.items()]
+        metadata_str = f"<metadata>{', '.join(metadata_items)}</metadata>"
+    return f"<document>{id_str}<content>{doc.page_content}</content>{metadata_str}</document>"

langchain-1.0.0a2/langchain/_internal/_lazy_import.py

@@ -0,0 +1,36 @@
+"""Lazy import utilities."""
+
+from importlib import import_module
+from typing import Union
+
+
+def import_attr(
+    attr_name: str,
+    module_name: Union[str, None],
+    package: Union[str, None],
+) -> object:
+    """Import an attribute from a module located in a package.
+
+    This utility function is used in custom __getattr__ methods within __init__.py
+    files to dynamically import attributes.
+
+    Args:
+        attr_name: The name of the attribute to import.
+        module_name: The name of the module to import from. If None, the attribute
+            is imported from the package itself.
+        package: The name of the package where the module is located.
+    """
+    if module_name == "__module__" or module_name is None:
+        try:
+            result = import_module(f".{attr_name}", package=package)
+        except ModuleNotFoundError:
+            msg = f"module '{package!r}' has no attribute {attr_name!r}"
+            raise AttributeError(msg) from None
+    else:
+        try:
+            module = import_module(f".{module_name}", package=package)
+        except ModuleNotFoundError as err:
+            msg = f"module '{package!r}.{module_name!r}' not found ({err})"
+            raise ImportError(msg) from None
+        result = getattr(module, attr_name)
+    return result

langchain-1.0.0a2/langchain/_internal/_prompts.py

@@ -0,0 +1,165 @@
+"""Internal prompt resolution utilities.
+
+This module provides utilities for resolving different types of prompt specifications
+into standardized message formats for language models. It supports both synchronous
+and asynchronous prompt resolution with automatic detection of callable types.
+
+The module is designed to handle common prompt patterns across LangChain components,
+particularly for summarization chains and other document processing workflows.
+
+"""
+
+from __future__ import annotations
+
+import inspect
+from typing import TYPE_CHECKING, Union
+
+if TYPE_CHECKING:
+    from collections.abc import Awaitable, Callable
+
+    from langchain_core.messages import MessageLikeRepresentation
+    from langgraph.runtime import Runtime
+
+    from langchain._internal._typing import ContextT, StateT
+
+
+def resolve_prompt(
+    prompt: Union[
+        str,
+        None,
+        Callable[[StateT, Runtime[ContextT]], list[MessageLikeRepresentation]],
+    ],
+    state: StateT,
+    runtime: Runtime[ContextT],
+    default_user_content: str,
+    default_system_content: str,
+) -> list[MessageLikeRepresentation]:
+    """Resolve a prompt specification into a list of messages.
+
+    Handles prompt resolution across different strategies. Supports callable functions,
+    string system messages, and None for default behavior.
+
+    Args:
+        prompt: The prompt specification to resolve. Can be:
+            - Callable: Function taking (state, runtime) returning message list.
+            - str: A system message string.
+            - None: Use the provided default system message.
+        state: Current state, passed to callable prompts.
+        runtime: LangGraph runtime instance, passed to callable prompts.
+        default_user_content: User content to include (e.g., document text).
+        default_system_content: Default system message when prompt is None.
+
+    Returns:
+        List of message dictionaries for language models, typically containing
+        a system message and user message with content.
+
+    Raises:
+        TypeError: If prompt type is not str, None, or callable.
+
+    Example:
+        ```python
+        def custom_prompt(state, runtime):
+            return [{"role": "system", "content": "Custom"}]
+
+        messages = resolve_prompt(custom_prompt, state, runtime, "content", "default")
+        messages = resolve_prompt("Custom system", state, runtime, "content", "default")
+        messages = resolve_prompt(None, state, runtime, "content", "Default")
+        ```
+
+    .. note::
+        Callable prompts have full control over message structure and content parameter
+        is ignored. String/None prompts create standard system + user structure.
+
+    """
+    if callable(prompt):
+        return prompt(state, runtime)
+    if isinstance(prompt, str):
+        system_msg = prompt
+    elif prompt is None:
+        system_msg = default_system_content
+    else:
+        msg = f"Invalid prompt type: {type(prompt)}. Expected str, None, or callable."
+        raise TypeError(msg)
+
+    return [
+        {"role": "system", "content": system_msg},
+        {"role": "user", "content": default_user_content},
+    ]
+
+
+async def aresolve_prompt(
+    prompt: Union[
+        str,
+        None,
+        Callable[[StateT, Runtime[ContextT]], list[MessageLikeRepresentation]],
+        Callable[[StateT, Runtime[ContextT]], Awaitable[list[MessageLikeRepresentation]]],
+    ],
+    state: StateT,
+    runtime: Runtime[ContextT],
+    default_user_content: str,
+    default_system_content: str,
+) -> list[MessageLikeRepresentation]:
+    """Async version of resolve_prompt supporting both sync and async callables.
+
+    Handles prompt resolution across different strategies. Supports sync/async callable
+    functions, string system messages, and None for default behavior.
+
+    Args:
+        prompt: The prompt specification to resolve. Can be:
+            - Callable (sync): Function taking (state, runtime) returning message list.
+            - Callable (async): Async function taking (state, runtime) returning
+              awaitable message list.
+            - str: A system message string.
+            - None: Use the provided default system message.
+        state: Current state, passed to callable prompts.
+        runtime: LangGraph runtime instance, passed to callable prompts.
+        default_user_content: User content to include (e.g., document text).
+        default_system_content: Default system message when prompt is None.
+
+    Returns:
+        List of message dictionaries for language models, typically containing
+        a system message and user message with content.
+
+    Raises:
+        TypeError: If prompt type is not str, None, or callable.
+
+    Example:
+        ```python
+        async def async_prompt(state, runtime):
+            return [{"role": "system", "content": "Async"}]
+
+        def sync_prompt(state, runtime):
+            return [{"role": "system", "content": "Sync"}]
+
+        messages = await aresolve_prompt(
+            async_prompt, state, runtime, "content", "default"
+        )
+        messages = await aresolve_prompt(
+            sync_prompt, state, runtime, "content", "default"
+        )
+        messages = await aresolve_prompt("Custom", state, runtime, "content", "default")
+        ```
+
+    .. note::
+        Callable prompts have full control over message structure and content parameter
+        is ignored. Automatically detects and handles async callables.
+
+    """
+    if callable(prompt):
+        result = prompt(state, runtime)
+        # Check if the result is awaitable (async function)
+        if inspect.isawaitable(result):
+            return await result
+        return result
+    if isinstance(prompt, str):
+        system_msg = prompt
+    elif prompt is None:
+        system_msg = default_system_content
+    else:
+        msg = f"Invalid prompt type: {type(prompt)}. Expected str, None, or callable."
+        raise TypeError(msg)
+
+    return [
+        {"role": "system", "content": system_msg},
+        {"role": "user", "content": default_user_content},
+    ]

langchain-1.0.0a2/langchain/_internal/_typing.py

@@ -0,0 +1,70 @@
+"""Private typing utilities for langchain."""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING, Any, ClassVar, Protocol, TypeAlias, TypeVar, Union
+
+from langgraph.graph._node import StateNode
+from pydantic import BaseModel
+
+if TYPE_CHECKING:
+    from dataclasses import Field
+
+
+class TypedDictLikeV1(Protocol):
+    """Protocol to represent types that behave like ``TypedDict``s.
+
+    Version 1: using ``ClassVar`` for keys.
+
+    """
+
+    __required_keys__: ClassVar[frozenset[str]]
+    __optional_keys__: ClassVar[frozenset[str]]
+
+
+class TypedDictLikeV2(Protocol):
+    """Protocol to represent types that behave like ``TypedDict``s.
+
+    Version 2: not using ``ClassVar`` for keys.
+
+    """
+
+    __required_keys__: frozenset[str]
+    __optional_keys__: frozenset[str]
+
+
+class DataclassLike(Protocol):
+    """Protocol to represent types that behave like dataclasses.
+
+    Inspired by the private ``_DataclassT`` from dataclasses that uses a similar
+    protocol as a bound.
+
+    """
+
+    __dataclass_fields__: ClassVar[dict[str, Field[Any]]]
+
+
+StateLike: TypeAlias = Union[TypedDictLikeV1, TypedDictLikeV2, DataclassLike, BaseModel]
+"""Type alias for state-like types.
+
+It can either be a ``TypedDict``, ``dataclass``, or Pydantic ``BaseModel``.
+
+.. note::
+    We cannot use either ``TypedDict`` or ``dataclass`` directly due to limitations in
+    type checking.
+
+"""
+
+StateT = TypeVar("StateT", bound=StateLike)
+"""Type variable used to represent the state in a graph."""
+
+ContextT = TypeVar("ContextT", bound=Union[StateLike, None])
+"""Type variable for context types."""
+
+
+__all__ = [
+    "ContextT",
+    "StateLike",
+    "StateNode",
+    "StateT",
+]

langchain-1.0.0a2/langchain/_internal/_utils.py

@@ -0,0 +1,7 @@
+# Re-exporting internal utilities from LangGraph for internal use in LangChain.
+# A different wrapper needs to be created for this purpose in LangChain.
+from langgraph._internal._runnable import RunnableCallable
+
+__all__ = [
+    "RunnableCallable",
+]

langchain-1.0.0a2/langchain/agents/__init__.py

@@ -0,0 +1,10 @@
+"""langgraph.prebuilt exposes a higher-level API for creating and executing agents and tools."""
+
+from langchain.agents.react_agent import AgentState, create_react_agent
+from langchain.agents.tool_node import ToolNode
+
+__all__ = [
+    "AgentState",
+    "ToolNode",
+    "create_react_agent",
+]

langchain-1.0.0a2/langchain/agents/_internal/__init__.py

@@ -0,0 +1 @@
+"""Internal utilities for agents."""

langchain-1.0.0a2/langchain/agents/_internal/_typing.py

@@ -0,0 +1,13 @@
+"""Typing utilities for agents."""
+
+from __future__ import annotations
+
+from collections.abc import Awaitable, Callable
+from typing import TypeVar, Union
+
+from typing_extensions import ParamSpec
+
+P = ParamSpec("P")
+R = TypeVar("R")
+
+SyncOrAsync = Callable[P, Union[R, Awaitable[R]]]

langchain-1.0.0a2/langchain/agents/interrupt.py

@@ -0,0 +1,92 @@
+"""Interrupt types to use with agent inbox like setups."""
+
+from typing import Literal, Union
+
+from typing_extensions import TypedDict
+
+
+class HumanInterruptConfig(TypedDict):
+    """Configuration that defines what actions are allowed for a human interrupt.
+
+    This controls the available interaction options when the graph is paused for human input.
+
+    Attributes:
+        allow_ignore: Whether the human can choose to ignore/skip the current step
+        allow_respond: Whether the human can provide a text response/feedback
+        allow_edit: Whether the human can edit the provided content/state
+        allow_accept: Whether the human can accept/approve the current state
+    """
+
+    allow_ignore: bool
+    allow_respond: bool
+    allow_edit: bool
+    allow_accept: bool
+
+
+class ActionRequest(TypedDict):
+    """Represents a request for human action within the graph execution.
+
+    Contains the action type and any associated arguments needed for the action.
+
+    Attributes:
+        action: The type or name of action being requested (e.g., "Approve XYZ action")
+        args: Key-value pairs of arguments needed for the action
+    """
+
+    action: str
+    args: dict
+
+
+class HumanInterrupt(TypedDict):
+    """Represents an interrupt triggered by the graph that requires human intervention.
+
+    This is passed to the `interrupt` function when execution is paused for human input.
+
+    Attributes:
+        action_request: The specific action being requested from the human
+        config: Configuration defining what actions are allowed
+        description: Optional detailed description of what input is needed
+
+    Example:
+        ```python
+        # Extract a tool call from the state and create an interrupt request
+        request = HumanInterrupt(
+            action_request=ActionRequest(
+                action="run_command",  # The action being requested
+                args={"command": "ls", "args": ["-l"]}  # Arguments for the action
+            ),
+            config=HumanInterruptConfig(
+                allow_ignore=True,    # Allow skipping this step
+                allow_respond=True,   # Allow text feedback
+                allow_edit=False,     # Don't allow editing
+                allow_accept=True     # Allow direct acceptance
+            ),
+            description="Please review the command before execution"
+        )
+        # Send the interrupt request and get the response
+        response = interrupt([request])[0]
+        ```
+    """
+
+    action_request: ActionRequest
+    config: HumanInterruptConfig
+    description: str | None
+
+
+class HumanResponse(TypedDict):
+    """The response provided by a human to an interrupt, which is returned when graph execution resumes.
+
+    Attributes:
+        type: The type of response:
+            - "accept": Approves the current state without changes
+            - "ignore": Skips/ignores the current step
+            - "response": Provides text feedback or instructions
+            - "edit": Modifies the current state/content
+        args: The response payload:
+            - None: For ignore/accept actions
+            - str: For text responses
+            - ActionRequest: For edit actions with updated content
+    """
+
+    type: Literal["accept", "ignore", "response", "edit"]
+    args: Union[None, str, ActionRequest]