mcp-omnisearch MCP Server

Use This MCP server To

Search the web and get sources I can click through Turn a long article into a short summary Get an answer with citations instead of ten blue links Find code examples across public GitHub projects Pull the readable text out of a cluttered page Compare what two search engines say about the same question

README

mcp-omnisearch

A Model Context Protocol (MCP) server that provides unified access to Tavily, Brave, Kagi, Exa AI, GitHub, Linkup, and Firecrawl through four consolidated tools.

Quick start

pnpm install
pnpm run build
node ./dist/index.js

Configure the server in your MCP client with whichever provider keys you have. Providers without keys are skipped and the rest keep working.

{
	"mcpServers": {
		"mcp-omnisearch": {
			"command": "node",
			"args": ["/path/to/mcp-omnisearch/dist/index.js"],
			"env": {
				"TAVILY_API_KEY": "your-tavily-key",
				"KAGI_API_KEY": "your-kagi-key",
				"BRAVE_API_KEY": "your-brave-key",
				"GITHUB_API_KEY": "your-github-token",
				"EXA_API_KEY": "your-exa-key",
				"LINKUP_API_KEY": "your-linkup-key",
				"FIRECRAWL_API_KEY": "your-firecrawl-key"
			}
		}
	}
}

Tools

`web_search`

Search the web with Tavily, Brave, Kagi, Exa, or Kagi Enrichment.

{
	"query": "sveltekit remote functions site:docs.svelte.dev",
	"provider": "brave",
	"limit": 10
}

`ai_search`

Get sourced AI answers with Kagi FastGPT, Exa Answer, or Linkup.

{
	"query": "Explain the differences between REST and GraphQL",
	"provider": "kagi_fastgpt"
}

`github_search`

Search GitHub code, repositories, or users.

{
	"query": "filename:remote.ts @sveltejs/kit",
	"search_type": "code",
	"limit": 5
}

`web_extract`

Extract, crawl, scrape, summarize, or find similar content with Tavily, Kagi, Firecrawl, or Exa.

{
	"url": "https://example.com/long-article",
	"provider": "kagi",
	"mode": "summarize"
}

Documentation

Provider selection — choose providers by task, key, mode, and capability.
Search operators — operator support matrix and tested examples.
Large results — inline vs file response behavior and remote deployment caveats.
Deployment — MCP client, WSL, Docker, cloud, and Firecrawl setup.
Troubleshooting — keys, access, validation, rate limits, and common failures.

Environment variables

TAVILY_API_KEY
KAGI_API_KEY
BRAVE_API_KEY
GITHUB_API_KEY
EXA_API_KEY
LINKUP_API_KEY
FIRECRAWL_API_KEY
FIRECRAWL_BASE_URL optional, for self-hosted Firecrawl
OMNISEARCH_LARGE_RESULT_MODE optional, file default or inline

Development

pnpm install
pnpm run build
pnpm test

Please read CONTRIBUTING.md before opening a PR.

License

MIT License - see LICENSE.

Acknowledgments

Built on Model Context Protocol, Tavily, Kagi, Brave Search, Exa AI, Linkup, and Firecrawl.

mcp-omnisearch FAQ

Is this still maintained?

Yes, very actively — the most recent commits landed in August 2026.

Do I need a key?

At least one. It accepts keys for Tavily, Brave, Kagi, Exa, Linkup, Firecrawl, and GitHub, and quietly skips any provider you have not set up.

Is it free?

The software is free, the search services generally are not. Several sell credits and a few offer a small free allowance, so check each one before you rely on it.

Can I use this to research a topic and get real sources?

Yes — that is its main job. You can ask for a search, a sourced answer, or a summary of one specific page.

Which apps does it work in?

Any assistant that can start a local MCP server; you point its config at the file you built.

How hard is setup?

It expects some comfort with a terminal: install the dependencies and build it once, then paste a config block holding your keys.