search-flights-browser

Extract flight offers from Google Flights pages into structured JSON.

1|Updated Apr 12, 2026
One-click install
npx skills add https://github.com/cryptopafi/nexusos-skills --skill search-flights-browser
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: search-flights-browser
Source: https://github.com/cryptopafi/nexusos-skills/tree/main/search-flights-browser
Command: npx skills add https://github.com/cryptopafi/nexusos-skills --skill search-flights-browser

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a robust browser-based fallback to locate flight offers when API-based MCP scrapers are blocked, rate-limited, or unavailable by navigating Google Flights (with Kayak as a fallback) and extracting structured results via AI page reading rather than fragile CSS selectors.

Core Features & Use Cases

  • Browser-based extraction: Uses agent-browser to navigate pre-filled Google Flights URLs or fill forms, wait for network idle, and capture the full page text for AI parsing.
  • AI page reading parser: Parses airline, numeric price, duration in minutes, stop count, departure and arrival times, and booking links from raw page text and returns results in the Travel Agent flight schema.
  • Fallback & error handling: Automatically switches to Kayak on CAPTCHA or blank results, supports partial results, currency notes, and explicit error responses for blocked sites or past dates. Example: use this Skill when Kiwi MCP and Google Flights MCP fail to return usable offers.

Quick Start

Use the search-flights-browser skill to find flights from OTP to LHR on 2025-12-30 in EUR returning up to 5 results.

Frequently Asked Questions about search-flights-browser

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract flight offer data when Google Flights API scrapers are blocked or rate-limited?

Browser-based flight search extraction navigates Google Flights with agent-browser, waits for network idle, and parses the rendered page text into structured JSON results sorted by price.

What is the best way to scrape flight prices from Google Flights without relying on fragile CSS selectors?

AI page reading parses airline, price, duration, stops, departure and arrival times directly from raw browser-rendered page text, avoiding CSS selector breakage and returning a structured flight schema.

How do I handle CAPTCHA or blank results when scraping Google Flights for one-way single-adult queries?

Automatic fallback switches to Kayak when Google Flights returns CAPTCHA or blank results, ensuring one-way single-adult flight searches still produce structured JSON offers.

Can I use agent-browser to retrieve full page text and parse flight data for past dates?

Agent-browser retrieves full page text for flight searches, but explicit error responses are returned for past dates, blocked sites, or unsupported queries rather than partial data.

Does this browser fallback approach return flight offers in a fixed schema with currency notes?

Yes, the parser returns results in a fixed Travel Agent flight schema sorted by price, supporting partial results, currency notes, and explicit error responses for blocked or unavailable sites.

Why use browser-rendered flight search extraction instead of API-based MCP scrapers for travel agent workflows?

Browser-rendered extraction provides a robust fallback when API-based MCP scrapers are unavailable, navigating pre-filled URLs and parsing full page text to locate flight offers reliably.