langchain-foxnose 0.1.0

LangChain integration for FoxNose - the knowledge layer for RAG and AI agents

langchain-foxnose

LangChain integration for FoxNose — the serverless knowledge platform purpose-built as the knowledge layer for RAG and AI agents.

Installation

pip install langchain-foxnose

Quick Start

from foxnose_sdk.flux import FluxClient
from foxnose_sdk.auth import SimpleKeyAuth
from langchain_foxnose import FoxNoseRetriever

# Create a FoxNose Flux client
client = FluxClient(
    base_url="https://<env_key>.fxns.io",
    api_prefix="my_api",
    auth=SimpleKeyAuth("YOUR_PUBLIC_KEY", "YOUR_SECRET_KEY"),
)

# Create the retriever
retriever = FoxNoseRetriever(
    client=client,
    folder_path="knowledge-base",
    page_content_field="body",
    search_mode="hybrid",
    top_k=5,
)

# Use it
docs = retriever.invoke("How do I reset my password?")
for doc in docs:
    print(doc.page_content)
    print(doc.metadata)

Features

• All search modes: text, vector, hybrid, and vector-boosted search
• Flexible content mapping: single field, multiple fields, or custom mapper function
• Metadata control: whitelist, blacklist, or include system metadata
• Native async: uses AsyncFluxClient for true async when available
• Structured filtering: pass FoxNose where filters for precise retrieval
• Full configuration: search fields, thresholds, hybrid weights, sort, and more

Search Modes

# Pure vector (semantic) search
retriever = FoxNoseRetriever(
    client=client,
    folder_path="articles",
    page_content_field="body",
    search_mode="vector",
)

# Hybrid search (text + vector)
retriever = FoxNoseRetriever(
    client=client,
    folder_path="articles",
    page_content_field="body",
    search_mode="hybrid",
    hybrid_config={"vector_weight": 0.6, "text_weight": 0.4},
)

# Text search with vector boost
retriever = FoxNoseRetriever(
    client=client,
    folder_path="articles",
    page_content_field="body",
    search_mode="vector_boosted",
    vector_boost_config={"boost_factor": 1.3},
)

Filtered Retrieval

retriever = FoxNoseRetriever(
    client=client,
    folder_path="articles",
    page_content_field="body",
    where={
        "$": {
            "all_of": [
                {"status__eq": "published"},
                {"category__in": ["tech", "science"]},
            ]
        }
    },
)

Async Usage

from foxnose_sdk.flux import AsyncFluxClient

async_client = AsyncFluxClient(
    base_url="https://<env_key>.fxns.io",
    api_prefix="my_api",
    auth=SimpleKeyAuth("YOUR_PUBLIC_KEY", "YOUR_SECRET_KEY"),
)

retriever = FoxNoseRetriever(
    async_client=async_client,
    folder_path="knowledge-base",
    page_content_field="body",
)

docs = await retriever.ainvoke("search query")

Documentation

• Getting Started
• Configuration
• Examples
• API Reference
• FoxNose Documentation

License

Apache-2.0 — see LICENSE for details.