The Universal Internet-to-App Engine

Fetch. Understand. Transform. Deliver.

Turn websites, documents, search results and the live web into clean, structured, AI-ready data—and send it anywhere.

TypeScript SDK

import { TayaproFetch } from "tayaprofetch";

const tf = new TayaproFetch({
  apiKey: process.env.TAYAPROFETCH_API_KEY,
});

const result = await tf.scrape({
  url: "https://example.com",
  formats: ["markdown", "json"],
});

console.log(result.content.markdown);

One platform, every retrieval surface

Every engine returns the same normalized document: content, metadata, links, provenance and cost — so downstream systems never change.

Scrape

One URL in, clean Markdown, text, metadata, links, images and tables out.

Crawl

Bounded breadth-first crawls with depth limits, dedup, budgets and a full site graph.

Map

Discover site architecture from sitemaps and links without fetching every document.

Extract

Schema or natural-language extraction into structured records with provenance.

Monitor

Content hashes and change detection that only fire your webhooks when it matters.

FetchFlow

Compose fetch, filter, transform and delivery steps into repeatable pipelines.

Compliant by default

robots.txt respected, SSRF protection, private networks and metadata endpoints blocked, per-domain policies.

One response model

success, content.markdown, content.structured, metadata, links, provenance, retrievedAt and contentHash on every document.

Model-agnostic AI

Structured extraction, summarisation and change detection behind a provider interface — never locked to one model vendor.

From sign-up to first fetch in five minutes

Create a workspace, generate a scoped API key, call /v1/scrape.

Start Building Try the Playground