# AnySearch

AnySearch — a hosted MCP server that gives an agent web search plus clean Markdown extraction. Handy when it needs a fact fast or has to read a page.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 841 installs
- Author: anysearch-ai
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/anysearch-ai/anysearch
- Updated: 2026-05-07
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/anysearch

## What it does

Four tools come with the server. `search` gives back ranked Markdown hits: title, link, snippet. `batch_search` takes between 2 and 5 unrelated queries, runs them together and merges the output into one reply so the context window stays smaller. `extract` pulls a page and returns its body as clean Markdown. `list_domains` hands over the sub-domain catalog plus the required query format for a domain.

It is a hosted HTTP endpoint at `https://anysearch.run.tools`, so nothing runs on your machine: you register the URL with your client. Search has two modes. Leave `domain` and `sub_domain` out for open web queries; pick them for a structured source, and call `list_domains` first, because the format rules it returns are mandatory.

The 2–5 query ceiling on `batch_search` is the one hard limit to plan around — longer lists need splitting. When a snippet is too short or the user wants a page read, the server tells the agent to follow up with `extract` instead of searching again.

## Requirements
- Remote server: `https://anysearch.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http anysearch https://anysearch.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add anysearch --url https://anysearch.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "anysearch": {
            "url": "https://anysearch.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "anysearch": {
            "type": "http",
            "url": "https://anysearch.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http anysearch https://anysearch.run.tools
```

## When to use it
- find fresh results for a query with links and snippets
- read a page by URL and get its text as Markdown
- run several queries in one call without filling the context
- check the required query format for a domain before searching
- verify a claim against the page it came from

## Example prompts
- Run `batch_search` with three queries about MCP HTTP transports and merge the answers.
- Call `list_domains` for finance first, then search the ticker in the required format.
- That snippet is cut off — open the page with `extract` and summarise the limits section.

## Similar
- [anysearch-mcp-server](https://affpapa.org/en/nevibecode/skills/anysearch-mcp-server.md) — skill · 226 installs
- [Keenable Web Search](https://affpapa.org/en/nevibecode/mcp/keenable-web-search.md) — MCP server · 2498 installs
- [firecrawl](https://affpapa.org/en/nevibecode/skills/firecrawl.md) — skill · 104500 installs
- [Web Search — Structured Results for AI Agents](https://affpapa.org/en/nevibecode/mcp/web-search-structured-results-for-ai-agents.md) — MCP server · 4334 installs
- [tavily-search](https://affpapa.org/en/nevibecode/skills/tavily-search.md) — skill · 39400 installs
- [OpenWebSearch](https://affpapa.org/en/nevibecode/mcp/openwebsearch.md) — MCP server · 494 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
