skills.rest
reason-machines
reason-machinesOFFICIAL
Explore all skills by this author →
Website𝕏@reasonmachinesGitHub

google-surf-mcp-search

Performs Google searches and extracts web and academic PDF content via an MCP server.

7|2|Updated May 16, 2026
Search Tags:#web scraping#pdf extraction#google search#academic papers#MCP server#parallel search#captcha recovery
One-click install
npx skills add https://github.com/reason-machines/mcp-skills --skill google-surf-mcp-search-reason-machines
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: google-surf-mcp-search
Source: https://github.com/reason-machines/mcp-skills/tree/main/skills/google-surf-mcp-search
Command: npx skills add https://github.com/reason-machines/mcp-skills --skill google-surf-mcp-search-reason-machines

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-surf-mcp.

What problem does it solve? Getting Google search results programmatically usually requires paid API keys, and scraping Google directly triggers CAPTCHAs and blocks. This Skill provides an MCP server that performs real Google searches without API keys, filters out ads and spam, and extracts content from web pages and academic PDFs. ## Core Features & Use Cases - Keyless Google Search: Run single or parallel Google searches with automatic ad, sponsored result, and knowledge panel filtering. - Content Extraction: Extract full text, abstracts, or metadata from HTML pages and PDFs, including academic sources like arXiv, Nature, and PubMed. - CAPTCHA Recovery: Automatically recover from CAPTCHAs using persistent browser profiles, with modes for local desktop, headless servers, and cloud environments. - Use Case: A research assistant agent searches for recent papers on a topic, triages results using token-efficient abstract mode, then extracts full text from the most promising papers for detailed analysis. ## Quick Start Add the google-surf MCP server to your MCP client config with npx -y google-surf-mcp, then ask your AI assistant to search Google for a topic and extract the top results.

Frequently Asked Questions about google-surf-mcp-search

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

FAQPage Schema
How do I search Google without an API key?▼

Use the google-surf-mcp server, which performs real Google searches through a browser automation layer instead of an API wrapper. Add it to your MCP client config with npx -y google-surf-mcp and call the search tool with a query and optional result limit.

How to extract text from academic PDFs like arXiv papers?▼

Call the extract tool with the PDF URL and choose a mode: full for complete text, abstract for a roughly 1500-character summary, or metadata for page count. It supports arXiv, bioRxiv, Nature, Springer, PubMed, and other academic sources, returning markdown-formatted text.

Does google-surf-mcp work on headless servers?▼

Yes, it supports headless servers through remote debugging mode. Set SURF_HEADLESS=true and SURF_REMOTE_DEBUG=true, then use SSH port forwarding and chrome://inspect to solve CAPTCHAs remotely when they appear.

Why am I getting repeated CAPTCHA requests?▼

CAPTCHA loops usually mean the browser profile is cold or the request rate is too high. Run npm run bootstrap to warm the profile, reduce SURF_RATE_LIMIT_PER_MIN to 5, and check the health tool to confirm the cascade mode is cycling properly.

What are the limitations of parallel search?▼

The search_parallel tool accepts a maximum of 10 queries per call using a worker pool. For larger batches, split queries into groups of 10 and add delays between batches to respect the built-in rate limiter, which defaults to 10 requests per minute.

Related Skills

Explore More Education & Research Skills→
hmzainjamilhmzainjamil
1

web-search

Formulate optimized web search queries and research strategies.

Community
Intermediate
hmzainjamilhmzainjamil
1

academic-search

Design academic literature search strategies and synthesize research findings from papers.

Community
Intermediate
blackplume233blackplume233
12

web-research

Conduct parallel web searches and extract cross-validated information from multiple sources.

Community
Intermediate
Arianna-LupiArianna-Lupi

configurar-serpapi

Configure and validate a SerpApi API key for live Google SERP data.

Official
Basic