Skip to main content

Changelog · September 28, 2026

Official LangChain and LlamaIndex packages

Pages an agent can read, and Documents ready to index, without writing the loader yourself.

By · Last updated: September 2026

TL;DR

Two packages on PyPI replace the loader recipes. langchain-urlpipe adds urlpipe_read_page, urlpipe_page_metadata, urlpipe_console_errors and urlpipe_lighthouse as LangChain tools, a UrlpipeToolkit and a UrlpipeLoader. llama-index-readers-urlpipe adds UrlpipeReader, with sync and async loading and a stable id per URL.

Free plan, no credit card. 1,000 credits a month.

Both build on the official Python client, so retries carry an Idempotency-Key, long pages are waited out, and errors come back typed. The LangChain tools return the API's explanation to the model when a page can't be read, so an agent can recover; the loaders skip failed pages with continue_on_failure.

Install
pip install langchain-urlpipe
pip install llama-index-readers-urlpipe

FAQ

Frequently asked questions

No. They use your project API key, from URLPIPE_API_KEY or passed in, and spend credits like any other call.

Try it on the free plan.

Free plan, no card. Confirm your email and your API key is live — you'll be making real requests in minutes.