MPPtempoquality 0.55

Extract clean content from up to 20 URLs via Tavily's AI-optimized extraction API, paid per-call on Tempo L2.

Price
per_call
Protocol
mpp
Verified
no
Endpoint
https://tavily.mpp.paywithlocus.com/tavily/extract

What it does

Tavily Extract is an MPP-enabled endpoint that pulls structured, clean content from one or more web pages. It is part of the broader Tavily suite (search, extract, map, crawl) proxied through the Locus MPP gateway and settled via Tempo L2 using pathUSD.

The endpoint accepts a POST request with a list of URLs (up to 20) and an optional extraction depth parameter ('basic' or 'advanced'). Pricing starts at $0.11 and scales with the number of URLs submitted. Payment is handled via the MPP charge intent — each call is a one-shot payment settled on-chain.

Note: The probe did not receive a 402 challenge on HEAD/GET (returned 404 instead), which is expected because this endpoint only responds to POST requests. The OpenAPI spec clearly documents the POST method and payment info, and the broader Tavily MPP service appears operational based on the well-formed OpenAPI manifest returned at the origin.

Capabilities

web-content-extractionurl-batch-processingbasic-and-advanced-depthai-optimized-extractionmpp-per-call-paymenttempo-l2-settlement

Use cases

  • —Extracting article text from a batch of news URLs for summarization pipelines
  • —Pulling product page content for comparison or monitoring agents
  • —Feeding clean web content into RAG retrieval systems
  • —Gathering research material from multiple sources in a single API call

Fit

Best for

  • —AI agents needing clean text from arbitrary web pages
  • —Batch extraction of up to 20 URLs in one request
  • —Pay-per-use content extraction without API key management

Not for

  • —High-volume scraping of thousands of pages per minute (20-URL cap per call)
  • —Extracting content from pages requiring authentication or JavaScript-heavy SPAs (not documented)

Quick start

curl -X POST https://tavily.mpp.paywithlocus.com/tavily/extract \
  -H "Content-Type: application/json" \
  -H "Authorization: <MPP-payment-header>" \
  -d '{"urls": "https://example.com/article1,https://example.com/article2", "extract_depth": "basic"}'

Example

Request

{
  "urls": "https://en.wikipedia.org/wiki/Artificial_intelligence,https://example.com/blog/post-1",
  "extract_depth": "basic"
}

Endpoint

URLhttps://tavily.mpp.paywithlocus.com/tavily/extract
Transporthttp
Protocolmpp
CurrencypathUSD

Quality

0.55/ 1.00

OpenAPI schema is complete with request body definitions and pricing metadata, but no 402 challenge was captured on HEAD/GET (expected since endpoint is POST-only). No example responses are available, and crawled pages only return 'not found' messages. Pricing is documented in x-payment-info description strings.

Warnings

  • —Probe did not capture a 402 MPP challenge because the endpoint only accepts POST; liveness not fully confirmed via probe
  • —No response schema or example response documented in the OpenAPI spec
  • —The 'urls' field is typed as string rather than array — likely expects comma-separated URLs but format is not explicitly documented

Citations

Provenance

Indexed frommpp_dev
Enriched2026-04-19 17:46:35Z · anthropic/claude-opus-4.6 · v2
First seen2026-04-18
Last seen2026-04-21

Agent access

JSONhttps://clawmart.sh/api/listings/TttkMz