🔗 Official Integration Guide

Give Your LangChain Agents Web Access

WebPerception API integrates with LangChain in 5 lines of code. Scrape any website, take screenshots, and extract structured data with AI.

Get Started in 3 Minutes

1 Install the SDK
pip install requests langchain
2 Create a LangChain Tool
from langchain.tools import Tool
import requests

def web_perceive(url: str) -> str:
    """Scrape a URL and extract structured data using AI."""
    response = requests.post(
        "https://api.mantisapi.com/v1/scrape",
        headers={"Authorization": "Bearer YOUR_API_KEY"},
        json={
            "url": url,
            "extract": {
                "schema": {
                    "title": "string",
                    "main_content": "string",
                    "links": ["string"]
                }
            }
        }
    )
    return str(response.json())

scrape_tool = Tool(
    name="web_scraper",
    description="Scrape a website and extract structured data. Input: URL",
    func=web_perceive
)
3 Add to Your Agent
from langchain.agents import initialize_agent, AgentType
from langchain.chat_models import ChatOpenAI

llm = ChatOpenAI(model="gpt-4", temperature=0)
agent = initialize_agent(
    tools=[scrape_tool],
    llm=llm,
    agent=AgentType.OPENAI_FUNCTIONS,
    verbose=True
)

# Your agent can now browse the web!
result = agent.run("What's the top story on Hacker News right now?")
print(result)

Why LangChain Developers Choose WebPerception

🧠

AI-Native Extraction

No CSS selectors. No XPath. Define a JSON schema and get structured data back. Works even when websites change their HTML.

⚡

JavaScript Rendering

Full browser rendering handles SPAs, infinite scroll, and dynamic content. Your agent sees exactly what a human sees.

🛡️

Anti-Bot Bypass

Built-in residential proxies, browser fingerprinting, and CAPTCHA handling. No blocked requests.

📸

Visual Perception

Take screenshots for your vision-enabled agents. Combine with GPT-4V for visual understanding of any webpage.

🔧

Tool-Ready

Designed as a LangChain Tool from day one. Works with OpenAI Functions, ReAct, and any agent architecture.

💰

Free Tier

100 requests/month free. No credit card required. Start building your web-enabled agent in minutes.

What LangChain + WebPerception Agents Can Do

🔍 Research Agent

Build an agent that researches any topic by crawling multiple sources, extracting key facts, and synthesizing a report.

result = agent.run(
    "Research the top 5 AI startups that raised Series A in 2026. "
    "For each, find their product, funding amount, and key investors."
)

📊 Competitor Monitor

Track competitor pricing, features, and messaging changes in real-time. Get alerts when something changes.

result = agent.run(
    "Check competitor.com/pricing and extract all plan names, "
    "prices, and included features. Compare with our current pricing."
)

📰 News Aggregator

Build an agent that monitors news sources and curates relevant articles for your team, complete with summaries.

result = agent.run(
    "Check TechCrunch, The Verge, and Hacker News for AI-related "
    "news from today. Summarize the top 3 stories."
)

Start Building Web-Enabled LangChain Agents

Free tier. No credit card. 100 requests/month to get started.

Get Free API Key → Read the Docs