AWS (亚马逊) 集成

from langchain_aws import ChatBedrock

from langchain_aws import ChatBedrockConverse

from langchain_aws import BedrockLLM

from langchain_aws import SagemakerEndpoint

from langchain_aws import BedrockEmbeddings

from langchain_aws.vectorstores.inmemorydb import InMemoryVectorStore

vds = InMemoryVectorStore.from_documents(
            chunks,
            embeddings,
            redis_url="rediss://cluster_endpoint:6379/ssl=True ssl_cert_reqs=none",
            vector_schema=vector_schema,
            index_name=INDEX_NAME,
        )

from langchain_aws.vectorstores import ValkeyVectorStore

from langchain_aws import AmazonKnowledgeBasesRetriever

from langchain_aws.tools import create_browser_toolkit

# 创建工具包
toolkit, browser_tools = create_browser_toolkit(region="us-west-2")

# 与智能体一起使用
agent = create_react_agent(model=llm, tools=browser_tools)
result = await agent.ainvoke(
    {"messages": [{"role": "user", "content": "Go to example.com and get the heading"}]},
    config={"configurable": {"thread_id": "session-1"}}
)

# 完成后清理
await toolkit.cleanup()

from langchain_aws.tools import create_code_interpreter_toolkit

# 创建工具包（异步）
toolkit, code_tools = await create_code_interpreter_toolkit(region="us-west-2")

# 与智能体一起使用
agent = create_react_agent(model=llm, tools=code_tools)
result = await agent.ainvoke(
    {"messages": [{"role": "user", "content": "Calculate factorial of 10"}]},
    config={"configurable": {"thread_id": "session-1"}}
)

# 完成后清理
await toolkit.cleanup()

from langchain_aws.graphs import NeptuneGraph
from langchain_aws.graphs import NeptuneAnalyticsGraph
from langchain_aws.chains import create_neptune_opencypher_qa_chain

from langchain_aws.graphs import NeptuneRdfGraph
from langchain_aws.chains import create_neptune_sparql_qa_chain

from langgraph_checkpoint_aws import AgentCoreMemorySaver

# 创建检查点保存器
checkpointer = AgentCoreMemorySaver(
    memory_id="your-memory-id",
    region_name="us-west-2"
)

# 与 LangGraph 一起使用
graph = workflow.compile(checkpointer=checkpointer)

# 使用 thread_id 和 actor_id 进行调用以实现对话持久化
config = {
    "configurable": {
        "thread_id": "user-123",
        "actor_id": "my-agent"  # AgentCore 必需
    }
}
result = graph.invoke({"messages": []}, config)

from langgraph_checkpoint_aws import AgentCoreMemoryStore

# 初始化用于长期记忆的存储
store = AgentCoreMemoryStore(memory_id="your-memory-id", region_name="us-west-2")

# 在模型前钩子中使用以保存和检索记忆
def pre_model_hook(state, config, *, store):
    actor_id = config["configurable"]["actor_id"]
    thread_id = config["configurable"]["thread_id"]
    namespace = (actor_id, thread_id)

    # 保存一条消息
    store.put(namespace, str(uuid.uuid4()), {"message": msg})

    # 搜索相关记忆
    results = store.search(("preferences", actor_id), query="user preferences", limit=5)
    return {"model_input_messages": state["messages"]}

from langchain_experimental.comprehend_moderation import AmazonComprehendModerationChain

from bedrock_agentcore.runtime import BedrockAgentCoreApp

app = BedrockAgentCoreApp()

@app.entrypoint
def agent_invocation(payload, context):
    result = graph.invoke({"messages": [{"role": "user", "content": payload["prompt"]}]})
    return {"result": result["messages"][-1].content}

app.run()

# Configure your agent
agentcore configure

# Deploy to AgentCore Runtime
agentcore launch -e your_agent.py