openserp

A self-hosted multi-engine search API supporting mainstream engines like Google and Bing.

MCPResearchOpen source

Type: MCP
Transport: http
Open source: Yes
GitHub Stars: ★ 1.5k
Source: mcp-github
Repository: github.com/karust/openserp

Overview

OpenSERP is a free and open-source search API and CLI tool that supports Google, Yandex, Baidu, Bing, DuckDuckGo, and Ecosia. No API keys or pay-per-use fees required—get real-time structured search results locally with just one command. It's suitable as a search tool for AI agents and LLMs, and can also serve as a backend for SEO rank tracking. Installable via Docker or Go, it offers multiple output formats including JSON and Markdown.

Capabilities

▪Supports six major search engines
▪Simultaneously queries across multiple engines with deduplication
▪Extracts clean Markdown content from target pages
▪Includes AI summaries, answer boxes, and other features
▪Supports image search and various filters

Use cases

Provides search capabilities for AI agentsSEO rank trackingAutomated data collection

Setup

Requires: DockerNode环境

Run via Docker: `docker run --rm -p 127.0.0.1:7000:7000 karust/openserp:latest serve -a 0.0.0.0 -p 7000` or install using Go: `go install github.com/karust/openserp@latest`

This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.

FAQ

Does it require an API key?

No, it's completely free and unlimited.

Which search engines are supported?

Google, Yandex, Baidu, Bing, DuckDuckGo, and Ecosia.

Related skills

Scrapling

An adaptive web scraping framework capable of handling everything from single requests to large-scale scraping.

MCP

TrendRadar

An AI-driven tool for sentiment monitoring and trend analysis, supporting multi-platform aggregation, RSS subscription, and intelligent notifications.

MCP