Episkey-G

Episkey-G / GrokSearch-rs

Public

Rust MCP server for Grok web search and Tavily-backed source retrieval

37
7
89% credibility
Found May 17, 2026 at 37 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

GrokSearch-rs is a free, open-source tool that gives your AI coding assistant the ability to search the internet and fetch web page content. Once connected, you can simply ask your assistant questions like 'find recent articles about Rust' and it will search the web, return a clear answer, and show you exactly where each piece of information came from. The tool works with Grok AI (from xAI) and can optionally use Tavily and Firecrawl to enrich search results with additional sources. It's designed to work seamlessly with popular AI coding tools like Claude Code, Cursor, and VS Code.

How It Works

1
🤔 You want your AI assistant to search the web

You're using an AI coding assistant and realize it can't browse the internet to find current information.

2
🔍 You discover GrokSearch-rs

You find this open-source tool that connects your AI assistant to real-time web search through Grok AI.

3
⚙️ You connect your accounts

You enter your Grok API key and Tavily API key so the tool can search the web and fetch page content.

4
✨ You link it to your AI assistant

With one simple command, you connect GrokSearch-rs to your AI coding tool so they can work together.

5
💬 You ask your assistant to search

You say something like 'search for the latest news on Rust programming' and your assistant springs into action.

6
The search happens automatically
📄
Grok AI finds the answer

Grok AI searches the web and returns a clear answer with citations to the sources it found.

🔗
Extra sources are added

Tavily finds additional relevant sources to make sure you get the most complete picture.

🎉 You get complete, sourced answers

Your assistant shows you the answer with clickable links to every source, so you can verify everything.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 37 to 37 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is GrokSearch-rs?

GrokSearch-rs is a Rust-based MCP server that brings live web search to AI assistants like Claude Code, Cursor, and VS Code. It connects to Grok's search API, enriches results with Tavily sources, and falls back to Firecrawl when needed. The server speaks JSON-RPC over stdio, so your AI client just launches it as a subprocess. You get five tools: web_search for sourced answers, get_sources to re-fetch cached results, web_fetch for page content, web_map for link discovery, and doctor for diagnostics. Configuration lives in a single TOML file, so multiple MCP clients share one set of API keys.

Why is it gaining traction?

The dual-transport design is the hook. Point it at xAI's native API or any OpenAI-compatible gateway by flipping environment variables. Source caching means repeated lookups hit memory instead of billing meters. Firecrawl as automatic fallback keeps fetch working when Tavily chokes. The doctor tool surfaces live connectivity probes with redacted config—useful when something breaks at 2am.

Who should use this?

Developers building AI-assisted workflows who need verified, sourced answers. DevOps teams wiring search into internal tooling. Anyone running multiple MCP clients who wants one config file instead of duplicating env blocks across tools. Researchers needing domain-filtered results with recency constraints.

Verdict

GrokSearch-rs is a solid, well-structured MCP server for search-heavy AI workflows. The dual-transport flexibility and source caching are genuinely useful. At 37 stars, it's early-stage but actively maintained. The 0.8999999761581421% credibility score reflects a small but engaged community. Worth trying if you need reliable web search in your AI pipeline, but expect some rough edges given the project's maturity.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.