anysearch-mcp-server

# AnySearch MCP Server Skill > Skill by [ara.so](https://ara.so) — MCP Skills collection. AnySearch MCP Server is a unified real-time search engine for AI agents, providing general web search, 23 vertical domain searches (finance, academic, security, legal, code, etc.), parallel batch search, and full-page URL content extraction via the MCP protocol. ## What It Does - **General Web Search**: Natural language queries across the web - **Vertical Domain Search**: Structured queries in 23 specialized domains - **Batch Search**: Execute up to 5 independent queries in parallel - **URL Extraction**: Fetch and convert full page content to Markdown - **Anonymous Access**: Works without API key (lower rate limits) ## Installation ### Streamable HTTP (Recommended - No Proxy) For agents supporting MCP spec 2025-03-26+ (OpenCode, Claude Desktop 2025.6+): **OpenCode** (`~/.opencode/config.json` or `opencode.json`): ```json { "mcp": { "anysearch": { "type": "streamable-http", "url": "https://api.anysearch.com/mcp", "headers": { "Authorization": "Bearer ${ANYSEARCH_API_KEY}" } } } } ``` **Claude Desktop** (`claude_desktop_config.json`): ```json { "mcpServers": { "anysearch": { "type": "streamable-http", "url": "https://api.anysearch.com/mcp", "headers": { "Authorization": "Bearer ${ANYSEARCH_API_KEY}" } } } } ``` ### stdio Transport (Via mcp-remote Proxy) **Claude Desktop** (`claude_desktop_config.json`): ```json { "mcpServers": { "anysearch": { "command": "npx", "args": [ "-y", "mcp-remote", "https://api.anysearch.com/mcp", "--header", "Authorization: Bearer ${ANYSEARCH_API_KEY}" ] } } } ``` **VS Code Copilot** (`.vscode/mcp.json`): ```json { "servers": { "anysearch": { "type": "stdio", "command": "npx", "args": [ "-y", "mcp-remote", "https://api.anysearch.com/mcp", "--header", "Authorization: Bearer ${ANYSEARCH_API_KEY}" ] } } } ``` **Cline** (VS Code settings): ```json { "mcpServers": { "anysearch": { "command": "npx", "args": [ "-y", "mcp-remote", "https://api.anysearch.com/mcp", "--header", "Authorization: Bearer ${ANYSEARCH_API_KEY}" ] } } } ``` ### SSE Transport (Via supergateway Proxy) Start the proxy server: ```bash npx -y supergateway \ --streamableHttp https://api.anysearch.com/mcp \ --outputTransport sse \ --port 8000 \ --oauth2Bearer ${ANYSEARCH_API_KEY} ``` **Cursor** (`.cursor/mcp.json`): ```json { "mcpServers": { "anysearch": { "type": "sse", "url": "http://localhost:8000/sse" } } } ``` **Windsurf** (`~/.codeium/windsurf/mcp_config.json`): ```json { "mcpServers": { "anysearch": { "serverUrl": "http://localhost:8000/sse" } } } ``` ### API Key Configuration **Optional but recommended**. Get a free key at: https://anysearch.com/console/api-keys Priority order: 1. `Authorization` header / `--api_key` flag 2. `ANYSEARCH_API_KEY` environment variable 3. `.env` file with `ANYSEARCH_API_KEY=<key>` 4. Anonymous access (lower rate limits) Set environment variable: ```bash export ANYSEARCH_API_KEY="your-api-key-here" ``` Or create `.env` file: ``` ANYSEARCH_API_KEY=your-api-key-here ``` ## Available Tools ### `search` - General or Vertical Domain Search Execute a search query. For vertical searches, you MUST call `list_domains` first. **Parameters:** | Parameter | Type | Required | Description | |-----------|------|----------|-------------| | `query` | string | Yes | Search query (format depends on domain) | | `domain` | string | No | Vertical domain (e.g., `finance`, `academic`) | | `sub_domain` | string | No | Sub-domain routing key (required for vertical) | | `sub_domain_params` | object | No | Extra params per sub_domain schema | | `content_types` | array | No | Filter: `web`, `news`, `code`, `doc`, etc. | | `zone` | string | No | `cn` or `intl` (required when sub_domain requires) | | `max_results` | integer | No | 1-100, default 10 | | `freshness` | string | No | `day`, `week`, `month`, `year` | **General Web Search Example:** ```json { "query": "quantum computing breakthroughs 2025", "max_results": 5, "freshness": "month" } ``` **Vertical Search Example (Stock):** First, call `list_domains`: ```json { "domain": "finance" } ``` Then use the discovered `sub_domain` and `query_format`: ```json { "query": "AAPL", "domain": "finance", "sub_domain": "finance.us_stock", "max_results": 5 } ``` **News Search Example:** ```json { "query": "artificial intelligence", "content_types": ["news"], "freshness": "week", "max_results": 10 } ``` **Academic Search Example:** ```json { "query": "10.1038/s41586-019-1666-5", "domain": "academic", "sub_domain": "academic.doi" } ``` ### `list_domains` - Query Vertical Domain Directory **MUST be called before vertical search** to discover available sub_domains and their query formats. **Parameters:** | Parameter | Type | Required | Description | |-----------|------|----------|-------------| | `domain` | string | One of | Single domain to query | | `domains` | array | One of | Batch up to 5 domains | **Single Domain Example:** ```json { "domain": "finance" } ``` **Multiple Domains Example:** ```json { "domains": ["finance", "academic", "security"] } ``` **Returns Markdown table with:** - `sub_domain`: Routing key for search - `description`: What the sub_domain searches - `query_format`: Exact format required (e.g., raw ticker, DOI) - `params_schema`: JSON schema for optional parameters - `zone`: If `CN`, must set `zone: "cn"` in search ### `batch_search` - Parallel Search Execution Execute 2-5 independent search queries in parallel. Single failure doesn't block others. **Parameters:** | Parameter | Type | Required | Description | |-----------|------|----------|-------------| | `queries` | array | Yes | 1-5 query objects, each with same fields as `search` | **Example:** ```json { "queries": [ { "query": "AAPL", "domain": "finance", "sub_domain": "finance.us_stock" }, { "query": "python async http client", "domain": "code", "sub_domain": "code.general" }, { "query": "CVE-2024-1234", "domain": "security", "sub_domain": "security.cve" } ] } ``` ### `extract` - URL Content Extraction Fetch full page content from a URL and return as Markdown. HTML pages only, truncated at 50,000 characters. **Parameters:** | Parameter | Type | Required | Description | |-----------|------|----------|-------------| | `url` | string | Yes | Target URL (`http://` or `https://`) | **Example:** ```json { "url": "https://en.wikipedia.org/wiki/Quantum_computing" } ``` ## Supported Domains **23 vertical domains available:** - `code` - Code repositories, packages, documentation - `tech` - Technology news, products, reviews - `fashion` - Fashion trends, brands, products - `travel` - Flights, hotels, destinations - `home` - Home improvement, furniture, decor - `ecommerce` - Products, prices, reviews - `gaming` - Games, platforms, news - `film` - Movies, reviews, showtimes - `music` - Songs, artists, albums - `finance` - Stocks, crypto, markets, companies - `academic` - Papers, patents, research - `legal` - Laws, regulations, cases - `business` - Companies, industries, markets - `ip` - Patents, trademarks, copyrights - `security` - CVEs, vulnerabilities, threats - `education` - Courses, schools, materials - `health` - Medical, wellness, drugs - `religion` - Religious texts, organizations - `geo` - Geographic data, maps, places - `environment` - Climate, conservation, ecology - `energy` - Energy sources, efficiency, policy - `ugc` - User-generated content, forums ## Common Usage Patterns ### Pattern 1: General Web Search When user asks an open-ended question: ```json { "query": "best practices for microservices architecture", "max_results": 10 } ``` ### Pattern 2: Recent News When user wants current information: ```json { "query": "SpaceX launches", "content_types": ["news"], "freshness": "week",