Extract structured content from URLs, paid per-URL via MPP/Tempo at $0.01 each.
What it does
The Parallel Extract endpoint (`POST /api/extract`) scrapes and extracts structured content from one or more web URLs. You supply an array of URLs and an optional `objective` string to guide what content is pulled. The endpoint is payment-gated via the Machine Payment Protocol (MPP) on the Tempo blockchain, settling in pathUSD at $0.01 per URL provided (minimum $0.01). The `npx mppx` CLI handles the HTTP 402 challenge, signing, and retry automatically.
This endpoint is part of the broader Parallel API gateway, which also offers web search (`/api/search`) and deep async research tasks (`/api/task`). All paid endpoints share a rate limit of 60 requests per minute per IP. The extract endpoint accepts a POST with a JSON body containing a required `urls` array and an optional `objective` field. No response schema is documented, but the endpoint returns extracted content for each URL upon successful payment.
Payment setup requires creating a Tempo account via `npx mppx account create` and funding it with pathUSD. All payments are final and non-refundable once confirmed on-chain. Free companion endpoints include `GET /api` for self-discovery of the full API schema, `GET /api/wallet/balance/:address` for checking your pathUSD balance, and `GET /api/task/:runId` for polling async task results.
Capabilities
Use cases
- —Extracting article text and key facts from news URLs for downstream summarization
- —Pulling structured product data from e-commerce pages
- —Feeding extracted web content into RAG pipelines or knowledge bases
- —Gathering competitive intelligence by extracting content from multiple competitor pages
- —Providing AI agents with clean page content instead of raw HTML
Fit
Best for
- —AI agents that need clean extracted web content on demand
- —Pipelines requiring structured extraction from arbitrary URLs
- —Developers who want pay-per-use extraction without API key management
- —Multi-URL batch extraction with optional objective-guided focus
Not for
- —High-volume crawling at thousands of URLs per minute (60 req/min rate limit)
- —Use cases requiring free or open-source extraction tooling
- —Scenarios where on-chain payment settlement is not acceptable
Quick start
# 1. Create a Tempo account
npx mppx account create
# 2. Fund wallet with pathUSD
# 3. Extract content from a URL
npx mppx https://parallelmpp.dev/api/extract --method POST -J '{"urls":["https://example.com"],"objective":"Extract key facts"}'Example
Request
{
"urls": [
"https://example.com",
"https://en.wikipedia.org/wiki/Web_scraping"
],
"objective": "Extract main article text and key facts"
}Endpoint
Quality
Full OpenAPI schema with request body, pricing, and examples are provided. The endpoint is part of a well-documented gateway with clear payment instructions. However, the probe did not capture a live 402 challenge (POST-only endpoint returned 404 on HEAD/GET, which is expected), and no response schema or example response is documented, preventing a higher score.
Warnings
- —Probe returned 404 on HEAD/GET because the endpoint only accepts POST — this is expected, not a sign the endpoint is down.
- —No response schema is documented for the 200 response; the structure of extracted content is unknown.
- —All payments are final and non-refundable once confirmed on-chain.
- —The /docs and /pricing pages return 404.
Citations
- —Extract endpoint is priced at $0.01 per URL (minimum $0.01)https://parallelmpp.dev
- —Payment is via MPP/Tempo using pathUSDhttps://parallelmpp.dev/api
- —Rate limit is 60 requests per minute per IPhttps://parallelmpp.dev
- —Request body requires a urls array and accepts an optional objective stringhttps://parallelmpp.dev/api
- —All payments are final with no refunds once confirmed on-chainhttps://parallelmpp.dev/api
- —npx mppx handles 402 challenge, signing, and retry automaticallyhttps://parallelmpp.dev