# срезAI — MCP-сервер и поисковый API для ИИ-агентов

> срезAI — MCP-сервер и поисковый API для ИИ-агентов is a paid MCP server for web search and scraping. Web search, page reading and YouTube subtitles for agents, priced per call in roubles. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Paid
- **Platforms:** API, CLI
- **Website:** https://srezai.ru/docs/mcp
- **Repository:** https://github.com/srezai-team/srezai-mcp-web-search
- **MCP Registry name:** ru.srezai/search
- **Categories:** web search and scraping
- **Listing updated:** 2026-11-14

## About

срезAI is a Russian search API for AI agents, and this server exposes it as 14 tools over streamable HTTP, authenticated with a srz_live key in the Authorization header. It covers the Runet and the wider web, and the documentation is in Russian with an English summary in the README.

Beyond search it reads pages into Markdown (long pages come back in windows via an offset), pulls only requested fields by schema, maps and crawls sites, and returns YouTube subtitles with timecodes without downloading video. Two tools hand over whole questions: answer_search ranks, reads and synthesises a sourced answer, and verify_claim returns a verdict with quotations from two to five independent domains.

One credit equals 0.10 rubles and a call costs the same regardless of how many results come back. Errors and empty calls are not charged. A small MIT-licensed npm bridge, srezai-mcp, is available for clients that only speak stdio.

## Features

- **Live web search:** Search with category, language, freshness and domain include or exclude lists.
- **Page to Markdown:** Reads one or many URLs, with offset windows for long documents.
- **Claim verification:** Returns a verdict with verbatim quotes from independent domains.
- **Schema extraction:** Returns only the fields you ask for.
- **YouTube subtitles:** Fetches manual or automatic captions with a from and to window.
- **Site mapping and crawl:** Lists up to 200 URLs and reads pages found.
- **Stdio bridge:** The srezai-mcp package forwards stdio clients to the same endpoint.

## Good for

- Give an agent search over Russian-language sources
- Fact-check a statement against several sites
- Read a long document in chunks that fit a context window
- Extract product fields from a list of pages

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http search https://srezai.ru/api/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://srezai.ru/api/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "search": {
      "url": "https://srezai.ru/api/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "search": {
      "type": "http",
      "url": "https://srezai.ru/api/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "search": {
      "serverUrl": "https://srezai.ru/api/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "search": {
      "type": "http",
      "url": "https://srezai.ru/api/mcp"
    }
  }
}
```

## Tools

- `web_search`: Live web search with categories, language, freshness and domain filters.
- `image_search`: Image search with direct links, source and resolution.
- `read_url`: Turns a page into clean Markdown, readable in windows by offset.
- `read_urls`: Reads up to five pages per call.
- `yt_srt`: Returns YouTube subtitles with timecodes.
- `fetch_page`: Renders a page in a browser and returns a screenshot and Markdown.
- `extract`: Pulls only the fields you specify from up to five links.
- `verify_claim`: Returns a supported, refuted, mixed or unverified verdict with quotes.
- `answer_search`: Returns a sourced answer after search, reranking and synthesis.
- `deep_research`: Runs multi-source agentic research and returns a ticket to collect the answer.
- `get_usage`: Shows balance, quota, spend and price list for free.
- `map`: Lists up to 200 site URLs from a sitemap or start page.
- `crawl`: Maps a site and returns text of the pages found.

## FAQ

### How is it priced?

One credit is 0.10 rubles; a web search, image search or page read costs one credit per call.

### Do I get free credit?

100 rubles arrive after Telegram confirmation, about 1,000 web requests, with no card needed.

### Do I need the npm package?

Only if your client supports stdio but not remote HTTP servers.

### Are failed calls billed?

No; errors and calls that return nothing are not charged.

### Is срезAI — MCP-сервер и поисковый API для ИИ-агентов free?

No. срезAI — MCP-сервер и поисковый API для ИИ-агентов is a paid product.

### How do I add срезAI — MCP-сервер и поисковый API для ИИ-агентов to Claude Code?

Run this command in your terminal: claude mcp add --transport http search https://srezai.ru/api/mcp

### What are the best alternatives to срезAI — MCP-сервер и поисковый API для ИИ-агентов?

The closest alternatives to срезAI — MCP-сервер и поисковый API для ИИ-агентов are Scrapling MCP Server, Firecrawl MCP Server, SearXNG Search, NeuralVerge and AI-Powered directory.

## Alternatives

- [Scrapling MCP Server](https://ai-powered.com/mcp/scrapling-mcp-server): Scrapling MCP Server is an open-source MCP server for web search and scraping.
- [Firecrawl MCP Server](https://ai-powered.com/mcp/firecrawl-mcp-server): Firecrawl MCP Server is a freemium MCP server for web search and scraping.
- [SearXNG Search](https://ai-powered.com/mcp/searxng-search): SearXNG Search is an open-source MCP server for web search and scraping.
- [NeuralVerge](https://ai-powered.com/mcp/neuralverge): NeuralVerge is an MCP server for web search and scraping.
- [AI-Powered directory](https://ai-powered.com/mcp/ai-powered-directory): AI-Powered directory is a free MCP server for developer tools.
- [Shopify MCP Server (any store)](https://ai-powered.com/mcp/shopify-mcp-server-any-store): Shopify MCP Server (any store) is a paid MCP server for e-commerce.
- [Chanroll: YouTube Channel Video Scraper (Official API)](https://ai-powered.com/mcp/chanroll-youtube-channel-video-scraper-official-api): Chanroll: YouTube Channel Video Scraper (Official API) is a paid MCP server for design and media.
- [Vidvett: YouTube Video Details Scraper (Official API)](https://ai-powered.com/mcp/vidvett-youtube-video-details-scraper-official-api): Vidvett: YouTube Video Details Scraper (Official API) is a paid MCP server for design and media.

---
Source: https://ai-powered.com/mcp/ai-mcp-api
