# LangGraph Prebuilt
This library defines high-level APIs for creating and executing LangGraph agents and tools.
## Installation
```bash
pip install langgraph-prebuilt
```
## Agents
`langgraph-prebuilt` provides an [implementation](https://langchain-ai.github.io/langgraph/reference/prebuilt/#langgraph.prebuilt.chat_agent_executor.create_react_agent) of a tool-calling [ReAct-style](https://langchain-ai.github.io/langgraph/concepts/agentic_concepts/#react-implementation) agent - `create_react_agent`:
```bash
pip install langchain-anthropic
```
```python
from langchain_anthropic import ChatAnthropic
from langgraph.prebuilt import create_react_agent
# Define the tools for the agent to use
def search(query: str):
"""Call to surf the web."""
# This is a placeholder, but don't tell the LLM that...
if "sf" in query.lower() or "san francisco" in query.lower():
return "It's 60 degrees and foggy."
return "It's 90 degrees and sunny."
tools = [search]
model = ChatAnthropic(model="claude-3-7-sonnet-latest")
app = create_react_agent(model, tools)
# run the agent
app.invoke(
{"messages": [{"role": "user", "content": "what is the weather in sf"}]},
)
```
## Tools
### ToolNode
`langgraph-prebuilt` provides an [implementation](https://langchain-ai.github.io/langgraph/reference/prebuilt/#langgraph.prebuilt.tool_node.ToolNode) of a node that executes tool calls - `ToolNode`:
```python
from langgraph.prebuilt import ToolNode
from langchain_core.messages import AIMessage
def search(query: str):
"""Call to surf the web."""
# This is a placeholder, but don't tell the LLM that...
if "sf" in query.lower() or "san francisco" in query.lower():
return "It's 60 degrees and foggy."
return "It's 90 degrees and sunny."
tool_node = ToolNode([search])
tool_calls = [{"name": "search", "args": {"query": "what is the weather in sf"}, "id": "1"}]
ai_message = AIMessage(content="", tool_calls=tool_calls)
# execute tool call
tool_node.invoke({"messages": [ai_message]})
```
### ValidationNode
`langgraph-prebuilt` provides an [implementation](https://langchain-ai.github.io/langgraph/reference/prebuilt/#langgraph.prebuilt.tool_validator.ValidationNode) of a node that validates tool calls against a pydantic schema - `ValidationNode`:
```python
from pydantic import BaseModel, field_validator
from langgraph.prebuilt import ValidationNode
from langchain_core.messages import AIMessage
class SelectNumber(BaseModel):
a: int
@field_validator("a")
def a_must_be_meaningful(cls, v):
if v != 37:
raise ValueError("Only 37 is allowed")
return v
validation_node = ValidationNode([SelectNumber])
validation_node.invoke({
"messages": [AIMessage("", tool_calls=[{"name": "SelectNumber", "args": {"a": 42}, "id": "1"}])]
})
```
## Agent Inbox
The library contains schemas for using the [Agent Inbox](https://github.com/langchain-ai/agent-inbox) with LangGraph agents. Learn more about how to use Agent Inbox [here](https://github.com/langchain-ai/agent-inbox#interrupts).
```python
from langgraph.types import interrupt
from langgraph.prebuilt.interrupt import HumanInterrupt, HumanResponse
def my_graph_function():
# Extract the last tool call from the `messages` field in the state
tool_call = state["messages"][-1].tool_calls[0]
# Create an interrupt
request: HumanInterrupt = {
"action_request": {
"action": tool_call['name'],
"args": tool_call['args']
},
"config": {
"allow_ignore": True,
"allow_respond": True,
"allow_edit": False,
"allow_accept": False
},
"description": _generate_email_markdown(state) # Generate a detailed markdown description.
}
# Send the interrupt request inside a list, and extract the first response
response = interrupt([request])[0]
if response['type'] == "response":
# Do something with the response
...
```
Raw data
{
"_id": null,
"home_page": "https://www.github.com/langchain-ai/langgraph",
"name": "langgraph-prebuilt",
"maintainer": null,
"docs_url": null,
"requires_python": "<4.0.0,>=3.9.0",
"maintainer_email": null,
"keywords": null,
"author": null,
"author_email": null,
"download_url": "https://files.pythonhosted.org/packages/9d/f0/5ac07b25d58a0a5bd126f4afc1a75ce4546f0e08fb4c3432154d904198d2/langgraph_prebuilt-0.1.0.tar.gz",
"platform": null,
"description": "# LangGraph Prebuilt\n\nThis library defines high-level APIs for creating and executing LangGraph agents and tools.\n\n## Installation\n\n```bash\npip install langgraph-prebuilt\n```\n\n## Agents\n\n`langgraph-prebuilt` provides an [implementation](https://langchain-ai.github.io/langgraph/reference/prebuilt/#langgraph.prebuilt.chat_agent_executor.create_react_agent) of a tool-calling [ReAct-style](https://langchain-ai.github.io/langgraph/concepts/agentic_concepts/#react-implementation) agent - `create_react_agent`:\n\n```bash\npip install langchain-anthropic\n```\n\n```python\nfrom langchain_anthropic import ChatAnthropic\nfrom langgraph.prebuilt import create_react_agent\n\n# Define the tools for the agent to use\ndef search(query: str):\n \"\"\"Call to surf the web.\"\"\"\n # This is a placeholder, but don't tell the LLM that...\n if \"sf\" in query.lower() or \"san francisco\" in query.lower():\n return \"It's 60 degrees and foggy.\"\n return \"It's 90 degrees and sunny.\"\n\ntools = [search]\nmodel = ChatAnthropic(model=\"claude-3-7-sonnet-latest\")\n\napp = create_react_agent(model, tools)\n# run the agent\napp.invoke(\n {\"messages\": [{\"role\": \"user\", \"content\": \"what is the weather in sf\"}]},\n)\n```\n\n## Tools\n\n### ToolNode\n\n`langgraph-prebuilt` provides an [implementation](https://langchain-ai.github.io/langgraph/reference/prebuilt/#langgraph.prebuilt.tool_node.ToolNode) of a node that executes tool calls - `ToolNode`:\n\n```python\nfrom langgraph.prebuilt import ToolNode\nfrom langchain_core.messages import AIMessage\n\ndef search(query: str):\n \"\"\"Call to surf the web.\"\"\"\n # This is a placeholder, but don't tell the LLM that...\n if \"sf\" in query.lower() or \"san francisco\" in query.lower():\n return \"It's 60 degrees and foggy.\"\n return \"It's 90 degrees and sunny.\"\n\ntool_node = ToolNode([search])\ntool_calls = [{\"name\": \"search\", \"args\": {\"query\": \"what is the weather in sf\"}, \"id\": \"1\"}]\nai_message = AIMessage(content=\"\", tool_calls=tool_calls)\n# execute tool call\ntool_node.invoke({\"messages\": [ai_message]})\n```\n\n### ValidationNode\n\n`langgraph-prebuilt` provides an [implementation](https://langchain-ai.github.io/langgraph/reference/prebuilt/#langgraph.prebuilt.tool_validator.ValidationNode) of a node that validates tool calls against a pydantic schema - `ValidationNode`:\n\n```python\nfrom pydantic import BaseModel, field_validator\nfrom langgraph.prebuilt import ValidationNode\nfrom langchain_core.messages import AIMessage\n\n\nclass SelectNumber(BaseModel):\n a: int\n\n @field_validator(\"a\")\n def a_must_be_meaningful(cls, v):\n if v != 37:\n raise ValueError(\"Only 37 is allowed\")\n return v\n\nvalidation_node = ValidationNode([SelectNumber])\nvalidation_node.invoke({\n \"messages\": [AIMessage(\"\", tool_calls=[{\"name\": \"SelectNumber\", \"args\": {\"a\": 42}, \"id\": \"1\"}])]\n})\n```\n\n## Agent Inbox\n\nThe library contains schemas for using the [Agent Inbox](https://github.com/langchain-ai/agent-inbox) with LangGraph agents. Learn more about how to use Agent Inbox [here](https://github.com/langchain-ai/agent-inbox#interrupts).\n\n```python\nfrom langgraph.types import interrupt\nfrom langgraph.prebuilt.interrupt import HumanInterrupt, HumanResponse\n\ndef my_graph_function():\n # Extract the last tool call from the `messages` field in the state\n tool_call = state[\"messages\"][-1].tool_calls[0]\n # Create an interrupt\n request: HumanInterrupt = {\n \"action_request\": {\n \"action\": tool_call['name'],\n \"args\": tool_call['args']\n },\n \"config\": {\n \"allow_ignore\": True,\n \"allow_respond\": True,\n \"allow_edit\": False,\n \"allow_accept\": False\n },\n \"description\": _generate_email_markdown(state) # Generate a detailed markdown description.\n }\n # Send the interrupt request inside a list, and extract the first response\n response = interrupt([request])[0]\n if response['type'] == \"response\":\n # Do something with the response\n ...\n```\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Library with high-level APIs for creating and executing LangGraph agents and tools.",
"version": "0.1.0",
"project_urls": {
"Homepage": "https://www.github.com/langchain-ai/langgraph",
"Repository": "https://www.github.com/langchain-ai/langgraph"
},
"split_keywords": [],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "255d8da7b0d22304ba128d57e6f691ab6c4c56b7a79f34c84a2f43a704539c70",
"md5": "b5c5789fd4698e1767c967f2bd08d802",
"sha256": "06ee27fd5ab6b860f2c44747ae81a1c814c5162ef8c4de6e4729d7b57ead788c"
},
"downloads": -1,
"filename": "langgraph_prebuilt-0.1.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "b5c5789fd4698e1767c967f2bd08d802",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<4.0.0,>=3.9.0",
"size": 24587,
"upload_time": "2025-02-27T00:03:13",
"upload_time_iso_8601": "2025-02-27T00:03:13.615390Z",
"url": "https://files.pythonhosted.org/packages/25/5d/8da7b0d22304ba128d57e6f691ab6c4c56b7a79f34c84a2f43a704539c70/langgraph_prebuilt-0.1.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "9df05ac07b25d58a0a5bd126f4afc1a75ce4546f0e08fb4c3432154d904198d2",
"md5": "38507b1eecbd90289a1412069a9109da",
"sha256": "b5acc0bd9dbebf66d738e1b31bbf75297f911de00de80a9b23f667342cd83988"
},
"downloads": -1,
"filename": "langgraph_prebuilt-0.1.0.tar.gz",
"has_sig": false,
"md5_digest": "38507b1eecbd90289a1412069a9109da",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<4.0.0,>=3.9.0",
"size": 23201,
"upload_time": "2025-02-27T00:03:15",
"upload_time_iso_8601": "2025-02-27T00:03:15.350645Z",
"url": "https://files.pythonhosted.org/packages/9d/f0/5ac07b25d58a0a5bd126f4afc1a75ce4546f0e08fb4c3432154d904198d2/langgraph_prebuilt-0.1.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-02-27 00:03:15",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "langchain-ai",
"github_project": "langgraph",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "langgraph-prebuilt"
}
JSON
