Gluedly Gluedly

· Product updates · Donatas · 2 min read

Introducing Native AI Framework Integrations: LangChain, LlamaIndex, and MCP Support

We are excited to announce a major milestone in Gluedly’s evolution as an LLM-ready web data extraction platform: official package releases across PyPI and npm!

Introducing Native AI Framework Integrations: LangChain, LlamaIndex, and MCP Support

As AI agents and RAG (Retrieval-Augmented Generation) pipelines move from prototype to production, feeding them clean, structured web data without HTML noise or anti-bot friction remains a significant hurdle. Gluedly was built from the ground up to solve this—mapping visual web elements into schema-stable JSON and prompt-ready Markdown.

Today, we are taking developer access a step further by launching native integrations for the most popular AI frameworks and protocols.

What’s New?

1. Model Context Protocol (MCP) Server for Claude & Cursor

You can now connect Gluedly directly to Claude Desktop, Cursor AI, and any MCP-compatible agent. This enables your AI assistants to inspect mapped pages, execute web scrapes, and retrieve snapshot data directly within your workspace environment.

npx @gluedly/mcp-server
  • Claude Desktop Setup: Add Gluedly to your claude_desktop_config.json:
{
  "mcpServers": {
    "gluedly": {
      "command": "npx",
      "args": ["-y", "@gluedly/mcp-server"],
      "env": {
        "GLUEDLY_API_KEY": "your_api_key_here"
      }
    }
  }
}

2. Official LangChain Integration

Building agentic workflows or document indexing pipelines with LangChain? The new langchain-gluedly Python package provides a seamless GluedlyLoader to transform page snapshots into LangChain Document objects automatically.

pip install langchain-gluedly
  • Usage:
from langchain_gluedly import GluedlyLoader

loader = GluedlyLoader(
    api_key="YOUR_GLUEDLY_API_KEY",
    page_id=12
)

# Automatically fetches latest snapshot and outputs ready-to-index Documents
documents = loader.load()

3. LlamaIndex Reader

For vector store ingestion and structured knowledge graphs, llama-index-readers-gluedly allows LlamaIndex users to ingest clean Markdown or JSON web data with minimal setup.

pip install llama-index-readers-gluedly
  • Usage:
from llama_index.readers.gluedly import GluedlyReader

reader = GluedlyReader(api_key="YOUR_GLUEDLY_API_KEY")
documents = reader.load_data(page_id=12)

Agentic SEO & Machine-Readable Docs

Alongside these library releases, we have deployed standard /llms.txt and /llms-full.txt endpoints at gluedly.com/llms.txt.

When web-enabled AI models or search agents (such as ChatGPT, Perplexity, or Claude) query the web for how to extract structured data for RAG or multi-site comparison tasks, Gluedly can now be discovered and integrated programmatically.

What's Next?

Our goal is to remain the premier web-data ingestion layer for AI stacks. Whether you are building autonomous research agents, e-commerce monitoring tools, or real-time RAG applications, you can now connect your web sources in minutes.