langchain-mixpeek 0.3.0

LangChain integration for Mixpeek — multimodal retriever and tool for AI agents

langchain-mixpeek

LangChain integration for Mixpeek — multimodal retriever and tool for searching video, image, audio, and document content from AI agents and RAG pipelines.

Install

pip install langchain-mixpeek

Quick start

Retriever (5 lines)

from langchain_mixpeek import MixpeekRetriever

retriever = MixpeekRetriever(
    api_key="mxp_...",
    retriever_id="ret_abc123",
    namespace="my-namespace",
)
docs = retriever.invoke("find the red cup")

Tool (5 lines)

from langchain_mixpeek import MixpeekTool

tool = MixpeekTool(
    api_key="mxp_...",
    retriever_id="ret_abc123",
    namespace="my-namespace",
)
result = tool.invoke("find the red cup")  # returns JSON string

Async retriever

from langchain_mixpeek import AsyncMixpeekRetriever

retriever = AsyncMixpeekRetriever(
    api_key="mxp_...",
    retriever_id="ret_abc123",
    namespace="my-namespace",
)
docs = await retriever.ainvoke("find the red cup")

Use in a chain

from langchain_core.prompts import ChatPromptTemplate
from langchain_openai import ChatOpenAI
from langchain_mixpeek import MixpeekRetriever

retriever = MixpeekRetriever(api_key="mxp_...", retriever_id="ret_...", namespace="ns")
llm = ChatOpenAI()

prompt = ChatPromptTemplate.from_template(
    "Answer using this context:\n{context}\n\nQuestion: {question}"
)

chain = (
    {"context": retriever, "question": lambda x: x}
    | prompt
    | llm
)
response = chain.invoke("what happens at 2 minutes?")

Use as an agent tool

from langchain.agents import AgentExecutor, create_openai_tools_agent
from langchain_openai import ChatOpenAI
from langchain_mixpeek import MixpeekTool

tools = [MixpeekTool(api_key="mxp_...", retriever_id="ret_...", namespace="ns")]
agent = create_openai_tools_agent(ChatOpenAI(), tools, prompt)
executor = AgentExecutor(agent=agent, tools=tools)
executor.invoke({"input": "What's in the video at the 2-minute mark?"})

Configuration

Parameter	Type	Default	Description
api_key	str	required	Mixpeek API key (mxp_...)
retriever_id	str	required	Mixpeek retriever ID (ret_...)
namespace	str	required	Mixpeek namespace to search
top_k	int	10 / 5	Max results (retriever / tool)
content_field	str	"transcript_chunk"	Metadata field used as page_content
filters	dict	None	Attribute filters passed to the retriever

Full docs

docs.mixpeek.com/agent-integrations/langchain