# mcp-omnisearch — Unified Search MCP Server
> Run mcp-omnisearch as an MCP server to unify Tavily, Brave, Kagi, Exa, GitHub search, and extraction tools behind one interface.
## Install
Merge the JSON below into your `.mcp.json`:
## Quick Use
```bash
git clone https://github.com/spences10/mcp-omnisearch && cd mcp-omnisearch
pnpm install
pnpm run build
# Configure your MCP client to run: node /path/to/mcp-omnisearch/dist/index.js
```
## Intro
Use mcp-omnisearch to give your agent a single MCP server for multi-provider web search and extraction workflows.
Run mcp-omnisearch as an MCP server to unify Tavily, Brave, Kagi, Exa, GitHub search, and extraction tools behind one interface.
**Best for:** agent builders who want one MCP endpoint that can fan out to multiple search/extraction providers with env-based keys
**Works with:** Node.js, pnpm, MCP clients that can run `node dist/index.js`, provider API keys via env vars
**Setup time:** ~10–20 minutes (install + build + client config)
### Key facts (verified)
- GitHub: 305 stars · 47 forks · pushed 2026-05-13.
- License: MIT · owner avatar + repo URL verified via GitHub API.
- Entrypoint (README-backed): `pnpm install && pnpm run build`.
## Main
- Start with one provider key enabled, then add more to avoid debugging multiple failures at once.
- Keep provider keys scoped and rotate them; treat search/extraction as an external dependency with quotas and costs.
- Add a thin policy layer in your agent: which providers are allowed for which task types and domains.
### README excerpt (source)
```text
# mcp-omnisearch
[](https://viteplus.dev)
[](https://vitest.dev)
A Model Context Protocol (MCP) server that provides unified access to
multiple search providers and AI tools. This server combines the
capabilities of Tavily, Brave, Kagi, Exa AI, GitHub, Linkup, and
Firecrawl to offer comprehensive search, AI responses, and content
processing through four consolidated tools.
## Features
### 🔍 `web_search` — Web Search
Search the web for information. Providers: tavily (factual/citations),
brave (privacy/operators), kagi (quality/operators), exa
(AI-semantic), kagi_enrichment (specialized indexes).
Parameters:
- `query` (string, required): Search query
- `provider` (string, required): `tavily`, `brave`, `kagi`, `exa`, or
`kagi_enrichment`
- `limit` (number, optional): Maximum number of results (default: 10)
```
### FAQ
**Q: Do I need all providers configured?**
A: No—enable only the providers you have keys for; the config is env-driven.
**Q: Where does it run?**
A: Your MCP client runs the Node entrypoint pointing at `dist/index.js` after build.
**Q: How do I keep results consistent?**
A: Prefer a primary provider per task type and log which provider served each answer.
## Source & Thanks
> Source: https://github.com/spences10/mcp-omnisearch
> License: MIT
> GitHub: 305 stars · 47 forks · pushed 2026-05-13
---
## 快速使用
```bash
git clone https://github.com/spences10/mcp-omnisearch && cd mcp-omnisearch
pnpm install
pnpm run build
# 在 MCP 客户端里配置：node /path/to/mcp-omnisearch/dist/index.js
```
## 简介
用 mcp-omnisearch 给 agent 提供一个统一的 MCP 服务入口，完成多数据源的搜索与抽取。
mcp-omnisearch 是聚合搜索的 MCP Server：把 Tavily、Brave、Kagi、Exa、GitHub 搜索与内容抽取服务统一到一个接口，便于在 agent 中复用与切换数据源。
**适合谁：** 希望用一个 MCP 端点统一对接多个搜索/抽取服务，并通过环境变量管理 key 的 agent 构建者
**可搭配：** Node.js、pnpm、可运行 `node dist/index.js` 的 MCP 客户端，通过环境变量配置各服务 API key
**上手时间：** 约 10–20 分钟（安装 + 构建 + 客户端配置）
### 关键事实（已验证）
- GitHub：305 stars · 47 forks · 最近更新 2026-05-13。
- 许可证：MIT（GitHub API 已验证仓库 URL / 头像等基础信息）。
- 入口命令（基于 README）：`pnpm install && pnpm run build`。
## 主要内容
- 先只启用一个 provider key 跑通链路，再逐步增加，避免一次调多个故障源。
- 把各 provider 的 key 当外部依赖来管理：最小权限、定期轮换，并关注配额/成本。
- 在 agent 侧加一层策略：不同任务/域名允许使用哪些 provider。
### README 摘录（来源）
```text
# mcp-omnisearch
[](https://viteplus.dev)
[](https://vitest.dev)
A Model Context Protocol (MCP) server that provides unified access to
multiple search providers and AI tools. This server combines the
capabilities of Tavily, Brave, Kagi, Exa AI, GitHub, Linkup, and
Firecrawl to offer comprehensive search, AI responses, and content
processing through four consolidated tools.
## Features
### 🔍 `web_search` — Web Search
Search the web for information. Providers: tavily (factual/citations),
brave (privacy/operators), kagi (quality/operators), exa
(AI-semantic), kagi_enrichment (specialized indexes).
Parameters:
- `query` (string, required): Search query
- `provider` (string, required): `tavily`, `brave`, `kagi`, `exa`, or
`kagi_enrichment`
- `limit` (number, optional): Maximum number of results (default: 10)
```
### FAQ
**Q: 必须配置所有 provider 吗？**
A: 不需要；只启用你有 key 的 provider，配置通过环境变量完成。
**Q: 在哪里运行？**
A: 由 MCP 客户端运行 Node 入口，构建后指向 `dist/index.js`。
**Q: 怎么保证结果一致？**
A: 按任务类型设定主 provider，并记录每次请求实际使用的 provider 便于回溯。
## 来源与感谢
> 来源：https://github.com/spences10/mcp-omnisearch
> 许可证：MIT
> GitHub：305 stars · 47 forks · 最近更新 2026-05-13
---
Source: https://tokrepo.com/en/workflows/mcp-omnisearch-unified-search-mcp-server
Author: MCP Hub