Web Scraper Skills
- Updated May 26, 2026
- OMIXEC/web-scraper-skills
web-scraper-skills is an open-source library of website-specific browser actions packaged as agent skills so coding agents can scrape and interact with sites programmatically without rediscovering selectors each run.
About
web-scraper-skills is an open-source library of website-specific browser action skills for AI agents. Each site exposes documented inputs, maintained JavaScript handlers, and structured JSON outputs so agents stop burning tokens on DOM inspection loops. Reach for it when you need programmatic access to sites that lack usable APIs, such as social feeds, marketplaces, or booking flows. It installs as per-site or umbrella skills for Claude Code, Cursor, Codex, and similar coding agents.
- Per-site skills for Reddit, X, Amazon, Booking.com, and more
- Actions run via page.evaluate or optional Chrome Bridge
- Generic scraper fallback for sites without prebuilt selectors
- Benchmarked lower token and time cost vs no-skill browsing
- Install via npx skills add or Claude marketplace manifest
Web Scraper Skills by the numbers
- Data as of Jul 7, 2026 (Skillselion catalog sync)
/plugin marketplace add OMIXEC/web-scraper-skillsAdd your badge
Show developers this marketplace is listed on Skillselion. Paste this into your README.
| Last updated | May 26, 2026 |
|---|---|
| Repository | OMIXEC/web-scraper-skills ↗ |
How can an AI agent interact with JavaScript-heavy websites reliably without endless DOM guessing, token waste, and brittle selector repair loops?
Give coding agents maintained, site-specific browser actions that return structured data without DOM guessing loops.
Who is it for?
Developers building agents that must read or act on social, commerce, or travel sites that do not expose the APIs they need.
Skip if: Teams that already have stable official APIs, simple static pages, or requirements that forbid browser automation under site terms.
What you get
The agent loads the matching site skill, runs a documented action, and receives structured JSON or display output with far fewer tokens and retries than manual browsing.
Plugins in this marketplace
1 plugin - install individually after you add the marketplace.
Recommended Marketplaces
FAQ
Do I need an external scraping API?
No. Skills run maintained JavaScript in a real browser context via page.evaluate or the optional chrome-bridge companion.
What if my target site is not listed?
Use the generic-scraper skill to extract text, links, tables, metadata, or CSV from any page without prebuilt selectors.
How do I install all supported sites at once?
Run npx skills add OMIXEC/Web Scraper Skills or point the agent at the umbrella SKILL.md in the repository.