# OpenSERP

OpenSERP adds live search across six engines, image search, and Markdown page extraction to AI coding clients. Useful when an agent needs sourced web results.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 3 repo stars
- Author: openserpapi
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/openserpapi/mcp
- Updated: 2026-10-27
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/openserp

## What it does

OpenSERP gives an AI agent tools for Google, Bing, Yandex, Baidu, DuckDuckGo, and Ecosia. Search responses retain the engine and original rank. The server also supports image queries, single-page extraction, and batch extraction, with separate tools for a preferred engine, fallbacks, and multi-engine comparison.

Run it with `npx -y @openserp/mcp` or install the package globally with npm. When `OPENSERP_API_KEY` is absent, it uses OSS mode at `http://localhost:7000`; Cloud access needs the API key. `batch_extract` accepts up to 20 URLs, while extracted pages are cut to context limits and can request more text through `maxChars`.

The server can be connected to Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, or Gemini CLI. Search will not work until a local OpenSERP service or a Cloud API key is available.

## Requirements
- Needs: Node.js (commands run via npx)
- Environment variables: `OPENSERP_API_KEY`, `OPENSERP_BASE_URL` — keys and tokens: ask the human for the values, never commit them or put them in shared files.

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add --env OPENSERP_API_KEY=<OPENSERP_API_KEY> --env OPENSERP_BASE_URL=<OPENSERP_BASE_URL> --transport stdio openserp -- npx -y @openserp/mcp
```
Replace <…> with your own values

### Codex

**Platform CLI**
```bash
codex mcp add openserp --env OPENSERP_API_KEY=<OPENSERP_API_KEY> --env OPENSERP_BASE_URL=<OPENSERP_BASE_URL> -- npx -y @openserp/mcp
```
Replace <…> with your own values

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "openserp": {
            "command": "npx",
            "args": [
                "-y",
                "@openserp/mcp"
            ],
            "env": {
                "OPENSERP_API_KEY": "<OPENSERP_API_KEY>",
                "OPENSERP_BASE_URL": "<OPENSERP_BASE_URL>"
            }
        }
    }
}
```
Replace <…> with your own values

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "openserp": {
            "type": "stdio",
            "command": "npx",
            "args": [
                "-y",
                "@openserp/mcp"
            ],
            "env": {
                "OPENSERP_API_KEY": "<OPENSERP_API_KEY>",
                "OPENSERP_BASE_URL": "<OPENSERP_BASE_URL>"
            }
        }
    }
}
```
Replace <…> with your own values

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add -e OPENSERP_API_KEY=<OPENSERP_API_KEY> -e OPENSERP_BASE_URL=<OPENSERP_BASE_URL> openserp npx -y @openserp/mcp
```
Replace <…> with your own values

## When to use it
- search across a specific engine
- compare Google and Bing rankings
- turn a web page into Markdown
- extract several URLs in one request
- check Cloud credit usage

## Example prompts
- Search Yandex for the current documentation for this library and return the first 5 URLs.
- Compare Google and Bing for this SEO query, keeping each result’s original rank.
- Extract these URLs as Markdown with `batch_extract` and use the pages to answer me.

## Similar
- [anysearch](https://affpapa.org/en/nevibecode/skills/anysearch.md) — skill · 38700 installs
- [find-the-original-image](https://affpapa.org/en/nevibecode/skills/find-the-original-image.md) — skill · 56600 installs
- [Google Lens](https://affpapa.org/en/nevibecode/mcp/google-lens.md) — MCP server · 387 installs
- [google-search](https://affpapa.org/en/nevibecode/skills/google-search.md) — skill · 4800 installs
- [SERPHouse](https://affpapa.org/en/nevibecode/mcp/serphouse.md) — MCP server · 107 installs
- [linkfox-amazon-search-by-image](https://affpapa.org/en/nevibecode/skills/linkfox-amazon-search-by-image.md) — skill · 410 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
