fetchary
v2.2.0
Published
Extract what matters. Semantic web extraction for Node.js and the command line, with optional browser login.
Downloads
1,156
Maintainers
Readme
Fetchary 👁️
Fetchary is a command-line tool for extracting information from web pages. It helps you collect text, links, contact details and public social metrics without manually searching each page. Choose what you need and get results ready to copy, save or use in another tool.
Start with a page URL and request a full analysis or a specific kind of information. You can turn a sitemap into a URL list, find contact details on a company page, or check the public metrics of a social profile. Save the results when you need to share them or continue your research later.
What it does
- Extract readable text, HTML or the contents of a selected page element.
- Find images, videos, favicons and profile pictures.
- Discover page links, sitemap URLs and downloadable documents.
- Collect email addresses, phone numbers and social profile links.
- Read available follower counts, post totals and engagement metrics.
- Inspect page metadata, structured data and public analytics or advertising tags.
- Read product prices and keep the evidence when page sources disagree.
- Combine eleven kinds of extraction in one page analysis report.
- Reuse signed-in sessions, save output and revisit recent commands.
See the social platform support matrix for available metrics and platform limitations.
Install
Use Fetchary on macOS, Linux or Windows with Node.js 26 or newer. Browser rendering and interactive sign-in also require the browser normally downloaded during installation. If it is missing, Fetchary prints the setup command to run; see browser setup.
Homebrew
brew install oliverjessner/tap/fetcharynpm
npm install --global fetcharyCLI
Extract specific information or analyse a page from your terminal:
fetchary analyse https://oliverjessner.at
fetchary extract text https://oliverjessner.at
fetchary extract links https://oliverjessner.at/sitemap.xml --listSee the CLI documentation for all commands and options.
Library
Use Fetchary in your own Node.js scripts to extract page content:
npm install fetcharyconst { extractText } = require('fetchary');
async function main() {
const result = await extractText('https://example.com');
console.log(result.data);
}
main().catch(console.error);See the library documentation for all methods and options.
Read pages and discover assets
Read a page as plain text, keep its HTML or select a specific element.
Find image and video URLs, embedded players and linked documents, or save a
favicon's original image with --binary.
When browser rendering finds a cookie banner, Fetchary tries to reject optional
cookies before reading the page. You can choose --consent accept explicitly or
disable this interaction with --consent off. See consent handling
for the available controls and limits.
Find contacts and social metrics
Collect email addresses, phone numbers and social profile links from a page. On supported social pages, read available follower counts, post totals and engagement metrics; unavailable values include a reason instead of becoming zero.
Check a product price
fetchary extract price https://example.com/product --jsonIf a page declares one price in its structured data and shows another in its product content, Fetchary returns both prices and where it found each one. A conflict leaves the selected price empty so your own tool can decide how to handle it. See the price reference for examples.
Inspect a page in one report
Use analyse to gather contacts, links, images, documents, metadata, structured
data, product prices, a favicon and public tracking integrations from one page.
Tracking IDs can help investigate connections between websites, but do not prove
common ownership.
Reuse signed-in sessions
Sign in through a browser with fetchary login and reuse the saved session for
later extraction. You can also supply an exported cookie file or, on macOS,
read the relevant cookies from Chrome; see signed-in sessions.
Save results and revisit commands
Add --filepath . to save output in the current directory, or choose a filename.
Use --list for values one per line, --json for structured results and
fetchary history to revisit your last 100 commands.
Who it is for
- Journalists and researchers gathering information from public web pages.
- Analysts checking company contacts, social profiles and available metrics.
- Content creators collecting image, video and document links.
- People automating repeated web research from the terminal.
Changelog
See the changelog for release notes and changes.
License
MIT
