#3333rd of 52 in MCP servers for developers

wigolo

Local MCP and REST server for web search, fetch, crawl and extraction by agents

Stars
5.5k
License
custom license
Last commit
Sep 2026
Last release
Jul 2026
Language
TypeScript
Stars, last 27 weeks+5,495
14 Apr 2026: 0 stars21 Apr 2026: 0 stars28 Apr 2026: 0 stars5 May 2026: 0 stars12 May 2026: 0 stars19 May 2026: 0 stars26 May 2026: 0 stars2 Jun 2026: 0 stars9 Jun 2026: 0 stars16 Jun 2026: 0 stars23 Jun 2026: 0 stars30 Jun 2026: 0 stars7 Jul 2026: 5 stars14 Jul 2026: 99 stars21 Jul 2026: 3,198 stars28 Jul 2026: 3,734 stars4 Aug 2026: 4,090 stars11 Aug 2026: 4,418 stars18 Aug 2026: 4,523 stars25 Aug 2026: 4,599 stars1 Sept 2026: 4,786 stars8 Sept 2026: 5,150 stars15 Sept 2026: 5,228 stars22 Sept 2026: 5,360 stars29 Sept 2026: 5,407 stars6 Oct 2026: 5,432 stars13 Oct 2026: 5,468 stars20 Oct 2026: 5,495 stars

0 on 14 Apr 20265,495 on 20 Oct 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

wigolo gives coding and self-hosted agents one tool surface for web search (18 engine adapters with rank fusion and on-device reranking), fetch, crawl, extract, cache, find-similar, research and an autonomous gather loop. It runs as an MCP server over stdio, a REST API on port 3333, or through TypeScript and Python SDKs. Core tools need no API keys; research and agent synthesis need an LLM provider or return a raw brief.

Who it is for: Developers giving coding or self-hosted agents local, keyless web access

Strengths

  • Search, fetch, crawl, extract and cache work without API keys; data stays under ~/.wigolo/
  • Results include verbatim excerpts with byte-offset source spans and a per-result score breakdown
  • Fetch escalates from plain HTTP to a headless browser on challenges or SPA shells
  • Also ships REST/OpenAPI, TS and Python SDKs, and LangChain, CrewAI, LlamaIndex, Vercel AI SDK wrappers

Weaknesses

  • Public beta status per the README
  • AGPL-3.0 per README badge; detected license is NOASSERTION, so check LICENSING.md
  • Synthesized answers from research and agent need an external or local LLM configured
  • Datacenter IPs may fail on challenge-protected sites; proxy is opt-in

What it needs

  • no GPU
  • Docker
  • Needs Node.js >= 20, ~1.5 GB disk for browser engine and models, Optional LLM provider (Gemini, Anthropic, OpenAI, Groq, Ollama)
  • Models: Gemini, Anthropic, OpenAI, Groq, Ollama
  • port 3333

Also in MCP servers for developers

See all 52
Also in MCP servers for developers
RankProjectScore
1gitnexusCode knowledge graph exposed to coding agents through MCP tools Live demo ↗ (opens in a new tab)48k starscustom license92 out of 100
2codegraphLocal code knowledge graph that coding agents query over MCP74.6k starsMIT82 out of 100
3n8n-mcpMCP server exposing n8n node docs, templates and workflow management to AI assistants Live demo ↗ (opens in a new tab)23.1k starsMIT77 out of 100
#4serenaMCP server giving coding agents symbol-level code search, editing and refactoring30.5k starscustom license74 out of 100
#5chrome-devtools-mcpMCP server that lets coding agents drive and debug a live Chrome53.9k starsApache-2.072 out of 100