!

Legal Disclaimer

PipeAgent is a data distribution gateway. We do not own, verify, or endorse the data provided by third-party creators. Use at your own discretion.

Docs/consumer / integration langchain

LangChain Integration Guide

LangChain is one of the most popular frameworks for building LLM applications. This guide shows you how to turn any pipeAgent feed into a tool that your LangChain Agent can use.

1. Prerequisites

  • Python 3.9+
  • pip install langchain langchain-openai requests
  • Your PipeAgent Read API Key (Found in your Dashboard).

    • 2. Basic Implementation

    The most efficient way to use PipeAgent in LangChain is via the @tool decorator.

    python
    import requests
from langchain.tools import tool
from langchain_openai import ChatOpenAI
from langchain.agents import create_openai_functions_agent, AgentExecutor
from langchain import hub

# Define the Tool
@tool
def product_search_tool(jsonpath: str = "$.data"):
    """
    Fetches real-time e-commerce data from PipeAgent. 
    Use this tool when users ask about product prices, availability, or ratings.
    """
    # Replace with your actual Feed ID
    FEED_ID = "your-feed-id-here"
    API_KEY = "your-read-api-key-here"
    
    url = f"https://api.pipeagent.dev/api/v1/feed/{FEED_ID}"
    headers = {"Authorization": f"Bearer {API_KEY}"}
    params = {"jsonpath": jsonpath}
    
    response = requests.get(url, headers=headers, params=params)
    return response.json()

# Setup the Agent
llm = ChatOpenAI(model="gpt-4o", temperature=0)
tools = [product_search_tool]
prompt = hub.pull("hwchase17/openai-functions-agent")

agent = create_openai_functions_agent(llm, tools, prompt)
agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)

# Run
agent_executor.invoke({"input": "What are the top 3 best selling mechanical keyboards right now?"})

    3. Advanced: Using JSONPath for Token Efficiency

    One of the biggest challenges with Agents is context window limits. PipeAgent's built-in JSONPath support allows you to filter data *before* it reaches the LLM.

    In your Tool definition, you can encourage the LLM to use specific paths:

    python
    @tool
def streamlined_search(query: str):
    """
    Search for items. The LLM will automatically construct a JSONPath 
    to fetch only the fields it needs (e.g. name and price).
    """
    # PipeAgent logic...
    pass

    4. Why Use PipeAgent with LangChain?

  • Zero Hallucination: Agents often hallucinate API structures. By providing the exact JSON Schema from the PipeAgent dashboard, you ensure the Agent knows exactly what fields exist.
  • Pre-cleaned Data: No need for complex BeautifulSoup or Selenium logic inside your LangChain chains.
  • Standardized Auth: One API key style across hundreds of different data sources.
    • Version 1.0.4 - Premium Infrastructure