Use case · News monitoring

New articles, into your knowledge base

Watch news sites, blogs and press rooms on a schedule. New articles are crawled into a knowledge base as clean markdown and announced by webhook, ready to search or summarise.

Start free View docs

30 min
checks, or daily, or any cron schedule
~500
tokens per article, on average
1B
free tokens on sign-up, no card

// For comms, research and media-intelligence teams

A PRESS ROOM, ON A SCHEDULE

press-watch.js · SarvCrawl Node SDKNode

import { KBClient } from 'sarvcrawl'
const kb = new KBClient()  // reads KB_API_KEY
// Crawl new articles from the press room every 30 minutes
await kb.monitor('kb_press', 'website', {
  url: 'https://news.example.com',
  include_paths: ['/press/*'],
  schedule: 'every 30 minutes',
  notify_url: 'https://hooks.example.com/new-articles',
})
// Later: what has been said about the merger?
const { hits } = await kb.embedSearchKb('kb_press', 'merger announcement')
0.90  Example Corp agrees to acquire Widget Co      /press/2026-10-12-acquisition
0.84  Regulators open review of proposed deal      /press/2026-10-13-review

Perfect for

Press and PR tracking

Follow the newsrooms of customers, partners and regulators and see every new release.

Industry news feeds

Collect trade publications and blogs into one searchable archive instead of a dozen RSS readers.

Newsletter and digest builders

Feed new articles to an LLM for summaries, with the source URL on every item.

Sites without feeds

Many press pages have no RSS. A website monitor finds new pages anyway.

How it works

01 / 06

Choose the sources

Point a website monitor at a news or press section, with include and exclude paths to keep it focused.

Map a site

02 / 06

Schedule the checks

Run every 30 minutes for breaking news, daily for a digest, or on any cron expression.

Monitor

03 / 06

Articles become markdown

Each new article is rendered in a real browser and stored as clean markdown with its URL and title.

Scrape

04 / 06

Get notified

Your webhook receives new, changed and removed pages, so your pipeline acts on fresh articles only.

n8n node

05 / 06

Search the archive

Search everything collected so far by keyword, or by meaning to catch stories that use other words (paid plans).

Keyword search Semantic search

06 / 06

Fill the gaps with search

Run a web search on a story and pull outside coverage into the same knowledge base.

How SarvCrawl compares

Feature	SarvCrawl	RSS readers	Google Alerts	Manual checks
Works on sites without a feed
Full article text stored
Your own schedule, down to minutes
Webhook for each new article
Searchable archive by meaning
Scriptable through an API

Turn press pages into a feed

Watch any news or press site and collect every new article. 1 billion free tokens, no card.

Start free View docs