
Build AI agents with live web data
About
Olostep is the most reliable and cost-effective web data API that lets companies, developers and AI agents scrape, crawl, and search the web at scale — returning clean markdown, screenshot, structured JSON, or AI-powered answers from any URL.
LangChain is an open-source Python and JavaScript framework for building LLM-powered applications, including agents, RAG pipelines, and multi-step AI workflows.
Capabilities
Extract clean markdown, HTML, or JSON from single URLs or autonomously crawl entire domains — with JS rendering and residential IPs included.
Submit up to 10,000 URLs in a single batch job. Results complete in 5–8 minutes, perfect for large-scale RAG ingestion.
Let your agent search the web and get structured answers with cited sources using the answer_question tool.
Getting started
Add Olostep to LangChain with a single config.
pip install langchain-olostep
# Then set your API key
export OLOSTEP_API_KEY="your_olostep_api_key_here"Get your API key from the Olostep dashboard. Free tier includes 500 requests — no credit card required.
Use Cases
Crawl documentation sites, blogs, or knowledge bases and ingest the results into your vector store for retrieval-augmented generation.
Build agents that look up companies, extract structured data from their websites, and enrich your CRM or spreadsheet automatically.
Combine extract_urls, scrape_batch, and answer_question into a LangGraph workflow that researches any topic end-to-end.
FAQ
How do I use Olostep with LangChain?
Install langchain-olostep via pip, set your OLOSTEP_API_KEY environment variable, and import the tools (scrape_website, answer_question, extract_urls, scrape_batch, crawl_website). Pass them to any LangChain agent or LangGraph workflow using the standard tools interface.
Does langchain-olostep work with LangGraph?
Yes. All Olostep LangChain tools are fully compatible with LangGraph. You can use them as nodes in a StateGraph workflow — for example, a discover → scrape → analyze pipeline that autonomously researches any topic.
What output formats does the LangChain integration support?
The scrape_website and scrape_batch tools support markdown, html, json, and text output formats. Markdown is recommended for LLM consumption, while JSON is useful for structured extraction pipelines.
Can I use Olostep tools with OpenAI function calling through LangChain?
Yes. Olostep tools are standard LangChain tool objects and work with any LangChain-compatible LLM, including OpenAI models that support function calling or tool use.
500 free credits to get started — no credit card required. Full documentation available in the Olostep docs.
← Back to all integrations