whatisgithub

What is groksearch-rs?

episkey-g/groksearch-rs — explained in plain English

Analysis updated 2026-05-18

★ 40RustAudience · developerComplexity · 3/5Setup · moderate

In one sentence

A lightweight Rust MCP server that gives AI coding assistants live web search using Grok, Tavily, and Firecrawl.

Mindmap

mindmap
  root((GrokSearch rs))
    What it does
      Live web search for AI tools
      MCP stdio server
    Tech Stack
      Rust
      Grok API
      Tavily
      Firecrawl
    Use Cases
      Cited web search
      Page content fetching
      Link discovery
    Audience
      Developers

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Give an AI coding assistant like Claude Code or Cursor live web search during a session.

USE CASE 2

Fetch full page content and cited sources for a research task inside your coding tool.

USE CASE 3

Discover links across a domain as part of an automated research workflow.

What is it built with?

RustMCPGrok APITavilyFirecrawl

How does it compare?

episkey-g/groksearch-rscesarferreira/tmux.exposeadysec/clawbot
Stars404137
LanguageRustRustRust
Setup difficultymoderateeasymoderate
Complexity3/52/53/5
Audiencedeveloperdeveloperdeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 30min

Needs an xAI Grok or OpenAI-compatible API key plus a Tavily key for search.

So what is it?

GrokSearch-rs is a lightweight server written in Rust that adds live web search capabilities to AI coding assistants. It works as an MCP stdio server, which means your AI tool (like Claude Code, Cursor, or VS Code) launches it in the background, and it quietly provides web search as a feature those tools can call upon. Under the hood, it connects to either xAI's Grok API or any OpenAI-compatible service to perform the search, then uses Tavily to fetch full page content and discover links, with Firecrawl as a backup. The tools it exposes include web search with cited sources, page content fetching, link discovery on a domain, and a diagnostic probe to check that everything is connected properly. Search results are cached so follow-up requests for sources are fast. You would use this when you want your AI assistant to access the live web during a coding or research session rather than being limited to its training data. Configuration is managed through environment variables or a shared config file, so multiple AI tools can share one set of API keys.

Copy-paste prompts

Prompt 1
Help me configure GrokSearch-rs as an MCP server for Cursor using environment variables.
Prompt 2
Explain how GrokSearch-rs falls back from Tavily to Firecrawl when fetching page content.
Prompt 3
Show me how to call GrokSearch-rs's web search tool with cited sources from my AI coding assistant.

Frequently asked questions

What is groksearch-rs?

A lightweight Rust MCP server that gives AI coding assistants live web search using Grok, Tavily, and Firecrawl.

What language is groksearch-rs written in?

Mainly Rust. The stack also includes Rust, MCP, Grok API.

How hard is groksearch-rs to set up?

Setup difficulty is rated moderate, with roughly 30min to a first successful run.

Who is groksearch-rs for?

Mainly developer.

Open on GitHub → Ask about another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.

What are similar repos?