Olostep Manifesto | read more →
Olostep
Free extraction tool

Turn any website into an API

Point Olostep at a public page, describe the fields you need, and get structured JSON plus a reusable /v1/scrapes call.

Website to API

Structured JSON + API snippet

Input
Output
Run the tool to see output here.
Named JSON fieldsReusable REST snippetNo CSS selectors

Developer experience

Turn any website into a structured API.

Use the Scrapes endpoint with llm_extract to describe the JSON fields you need, and get structured data from any public URL.

REST APIJSON outputAI extractionCustom promptsCountry targetingAny public URL
Read the docs
curl --request POST \
  --url https://api.olostep.com/v1/scrapes \
  --header 'Authorization: Bearer YOUR_API_KEY' \
  --header 'Content-Type: application/json' \
  --data '{
    "url_to_scrape": "https://example.com/product",
    "formats": ["markdown","json"],
    "wait_before_scraping": 0,
    "llm_extract": {
      "prompt": "Extract product name, price, description, and features"
    },
    "country": "US"
  }'

Extract the fields you actually need

Most scrapers hand back the whole page. This tool asks Olostep to return only the structured fields you describe, then shows the same call you can run from your stack.

  • Describe fields, not selectors

    Write a prompt such as product name, price, and availability. Olostep renders the page and returns named JSON.

  • The extract is an API call

    Copy the /v1/scrapes snippet with the same URL and llm_extract prompt. Call it from curl, Node, or any language.

  • Scale with parsers and batches

    This free check is one page. For catalogs, schedules, or site-specific fields, use parsers, batches, and monitors.

How it works

Each tool is designed around a direct job: provide the input, get a useful output, and move it into your workflow.

  • 1

    Enter a public URL

    Use any live page. A scheme is added automatically if you leave it off.

  • 2

    Describe the JSON you want

    Leave the default prompt, or name the fields. Olostep extracts them with llm_extract.

  • 3

    Copy JSON or the API snippet

    Review the structured result, then paste the same /v1/scrapes call into your app.

Built for practical workflows

These tools are small on purpose: they solve the conversion job quickly, then point teams toward the API path when the workflow becomes recurring or high volume.

  • Product and pricing pages

    Pull name, price, stock, and image URLs into a feed without writing a parser first.

  • Listings and directories

    Prototype a structured extract from a search or listing page before you move it to batches.

  • Content for RAG and agents

    Get typed fields instead of a markdown dump so downstream models do not have to clean the page.

Need this at scale?

Run the same extract in the playground or call the Scrapes API. For catalogs and schedules, use parsers, batches, and monitors.

Frequently asked questions

Product & Capabilities

  • Olostep scrapes the public URL, extracts the fields you described with llm_extract, and returns structured JSON. The same payload is a POST to /v1/scrapes you can call with your API key.

  • HTML to JSON parses markup you paste in the browser. This tool fetches a live URL and extracts named fields. It does not require the HTML source.

  • URL to JSON is the product landing page with the full scrape playground. This page is a focused free check: URL, fields, JSON, and a copyable API snippet.

Usage & Automation

  • It is a parameter on POST /v1/scrapes. You pass a natural-language prompt and/or a JSON schema. Olostep returns the match in result.json_content.

  • Yes. Olostep renders the page before extraction, so JS content is included. Very slow pages can take a minute or two.

  • No. This free check does not persist a trained agent. Reuse the snippet, or create a parser in the dashboard for a durable, site-specific extract.

Pricing & Plans

  • The page UI is free to use. Live extracts use the Olostep API configured for this site and are rate limited.

  • A standard scrape is one credit. llm_extract uses additional credits because of the model step. See Olostep pricing for current rates.

  • Olostep offers free credits for new users, so teams can test extraction before scaling.

Website to API | Turn a Page into JSON