LangChain

Getting started

LangChain is the most widely used framework for building LLM-powered agents. The langchain-agentmail package wraps the AgentMail SDK as standard LangChain tools, plus a document loader and a retriever — so a LangGraph agent can send, reply, draft, label, and search email through a real inbox without any glue code.

The full reference lives on the LangChain docs site: docs.langchain.com/oss/python/integrations/providers/agentmail.

Use cases

• Give agents their own inboxes: Provision a dedicated email address per agent so it can send and receive mail independently.
• Triage and reply: Read recent threads, summarize what’s new, and reply inside the same thread with the right In-Reply-To headers.
• Stage and schedule sends: Use the draft tools to compose iteratively or schedule a delivery time via send_at.
• RAG over email: Load messages as LangChain Documents and index them into a vector store for semantic search across the inbox.

Prerequisites

1. An AgentMail account with an API key from the AgentMail Console.
2. Python 3.10+ and a LangChain-compatible model provider (e.g. an OPENAI_API_KEY or ANTHROPIC_API_KEY).

Setup

Install the integration package:

$
pip install langchain-agentmail

Set your API key:

$
export AGENTMAIL_API_KEY="your-api-key"

Quickstart

Build a ReAct agent with the full toolkit in a few lines:

1
from langchain_openai import ChatOpenAI
2
from langgraph.prebuilt import create_react_agent
3
4
from langchain_agentmail import AgentMailToolkit
5
6
toolkit = AgentMailToolkit.from_api_key()
7
agent = create_react_agent(
8
    ChatOpenAI(model="gpt-4o-mini"),
9
    tools=toolkit.get_tools(),
10
)
11
12
result = agent.invoke(
13
    {"messages": [("user", "Summarize my most recent email thread.")]}
14
)
15
print(result["messages"][-1].content)

You can also pull a single tool in if you don’t need the whole toolkit:

1
from langchain_agentmail import AgentMailClient, AgentMailSendTool
2
3
send = AgentMailSendTool(client=AgentMailClient())
4
send.invoke({
5
    "inbox_id": "ib_...",
6
    "to": "alice@example.com",
7
    "subject": "Ping",
8
    "text": "Hello from my agent.",
9
})

Available tools

The toolkit exposes one tool per AgentMail operation.

Inbox and thread management

Message operations

Draft management

RAG over an inbox

AgentMailLoader streams messages as LangChain Documents — one per message, plain-text body as page_content, sender / subject / labels / thread / attachment metadata on metadata. Pair it with any vector store for semantic search:

1
from langchain_core.vectorstores import InMemoryVectorStore
2
from langchain_openai import OpenAIEmbeddings
3
4
from langchain_agentmail import AgentMailLoader
5
6
docs = AgentMailLoader(inbox_id="ib_...", limit=200).load()
7
store = InMemoryVectorStore.from_documents(docs, OpenAIEmbeddings())
8
retriever = store.as_retriever(search_kwargs={"k": 5})
9
10
retriever.invoke("Q3 invoice from acme")

For a quick keyword search without embeddings, use the bundled AgentMailRetriever instead.

Inbound email via webhooks

The webhooks extra ships a FastAPI router with svix-compatible signature verification so a LangGraph agent can react to inbound mail:

$
pip install 'langchain-agentmail[webhooks]'

1
from fastapi import FastAPI
2
from langchain_agentmail.webhooks import AgentMailEvent, create_fastapi_router
3
4
async def on_event(event: AgentMailEvent) -> None:
5
    if event.event_type == "message.received":
6
        # drive your LangGraph agent here
7
        ...
8
9
app = FastAPI()
10
app.include_router(
11
    create_fastapi_router(on_event),  # reads AGENTMAIL_WEBHOOK_SECRET
12
    prefix="/agentmail",
13
)

Resources

• LangChain integration docs: docs.langchain.com/oss/python/integrations/providers/agentmail
• Source code: github.com/agentmail-to/langchain-agentmail
• PyPI: pypi.org/project/langchain-agentmail/