Skip to main content
MCP server CX Codex

Geekflare · geekflare — Claude Code MCP server: what it does, how to install

An MCP server that connects Geekflare API to Claude Code and other AI assistants. Scrapes pages, pulls meta tags, gathers brand data and SERP.

MIT ★ 1 updated 2026-10-22

Install into Claude Code · Platform CLI All methods and platforms ↓
claude mcp add --env API_KEY=<API_KEY> --transport stdio geekflare-geekflare -- npx -y @geekflare/mcp
Open on GitHub

What Geekflare does

Geekflare brings web intelligence tools into the editor. Three core tools: webScrape fetches page content as HTML, Markdown, JSON or LLM-optimised formats, extracts structured data via CSS/XPath schemas or ready templates (product, contact), renders JavaScript and routes through country-specific proxies. metaScrape pulls meta tags, Open Graph and Twitter cards. brand collects brand identity: logos, colors, typography, social profiles and company details.

Requires an API key from geekflare.com/api. The server runs via npx or Docker, key passed as API_KEY environment variable. HTTP transport is available at mcp.geekflare.com/{API_KEY}/mcp if stdio is not an option. AI extraction (via aiPrompt parameter in webScrape) adds 6 credits on top of base request cost.

All three tools block ads by default and bypass CAPTCHAs with stealth mode. For webScrape, you can set a post-load delay (waitTime) and return a file instead of text (fileOutput).

What it helps with

  • fetch page content in Markdown for LLM input
  • extract product data or contacts via CSS schema
  • pull Open Graph and meta tags for SEO audit
  • collect brand assets: logos, colors, social links
  • scrape through a country-specific proxy

Example prompts for the agent

  • Pull all meta tags and Open Graph from example.com
  • Scrape shop.example.com/product/123 as JSON using the product template
  • Collect brand data for company.example.com, including typography and social profiles

How to install Geekflare

1Platform CLI

claude mcp add --env API_KEY=<API_KEY> --transport stdio geekflare-geekflare -- npx -y @geekflare/mcp

Replace <…> with your own values

2Platform CLI: remote server

claude mcp add --transport http mcp https://mcp.geekflare.com/{API_KEY}/mcp

Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more

Questions about Geekflare

How to install Geekflare in Claude Code?

Run claude mcp add --env API_KEY=<API_KEY> --transport stdio geekflare-geekflare -- npx -y @geekflare/mcp, where <API_KEY> is your key from geekflare.com/api. Or via HTTP: claude mcp add --transport http geekflare-geekflare https://mcp.geekflare.com/{API_KEY}/mcp.

What does Geekflare do?

Scrapes pages into Markdown, HTML or JSON, extracts meta tags and Open Graph, collects brand data: logos, colors, typography and social profiles. Renders JavaScript, extracts data via custom schemas and works through proxies.

What extraction formats does webScrape support?

html, markdown, json, markdown-llm, html-llm, text, text-llm. For JSON, there are ready templates (product, contact), custom CSS/XPath schemas and AI extraction via aiPrompt parameter.

Similar MCP servers

All MCP servers for Browser and scraping →

Geekflare by other authors comes first.

Geekflare hello-jek2 Geekflare brings web scraping, network diagnostics, and PDF generation to AI agents. Helpful for website audits and remote data collection. CX Codex +1 586installs search brightdata Routes agent queries to Google, Bing, or answer engines like Perplexity. Useful when work starts from an open-ended search query rather than a known URL. CX Codex +1 7,700installs tracking-google-serp-rankings apidojo tracking-google-serp-rankings is a Claude Code skill that collects Google rankings and SERP features for search terms. It runs for competitor checks, snippet research, and keyword audits. CX Codex +1 922installs scraping-google-search-results apidojo scraping-google-search-results — a Claude Code skill that runs the apidojo Google scraper on Apify and returns SERP data for any query. Covers organic results, featured snippets, and People Also Ask boxes. CX Codex +1 917installs HasData DuckDuckGo HasData HasData DuckDuckGo supplies structured DuckDuckGo search output in JSON. Ideal for scraping organic rankings, inspecting ad blocks, and checking 37 regional markets. CX Codex +1 ★ 6repo stars SERP Maven nodemaven The SERP Maven MCP server for Claude Code connects AI agents to live Google search results through a real browser and residential proxies. It runs search queries from the IDE. CX Codex +1 ★ 129repo stars bright-data-best-practices brightdata bright-data-best-practices — a reference skill that teaches a coding assistant which Bright Data API fits the task and how to wire it up. Handy when the agent builds a scraper or a bdata wrapper. CX Codex +1 1,800installs

Project sponsors

Companies that keep this analytics open