Agent Web Search MCP Server

Agent Web Search is an MCP integration. Provider-neutral web_search for AI agents. Grounding + agent search APIs, keyless defaults.

Packages and local setup

The published pypi package is agent-web-search-mcp, version 0.8.0. Its declared MCP transport is stdio.

The package declares AGENT_WEB_SEARCH_PROVIDERS: Comma-separated startup-enabled provider set; defaults to ddgs,exa,parallel (all keyless); ARK_API_KEY: Volcengine Ark API key for Doubao grounding; GEMINI_API_KEY: Google Gemini API key for Google Search grounding; XAI_API_KEY: xAI API key for Grok web/X search; DEEPSEEK_API_KEY: DeepSeek API key for native web search; AGENT_WEB_SEARCH_RESPONSES_API_KEY: Generic Responses API key for web search grounding (falls back to OPENAI_API_KEY); AGENT_WEB_SEARCH_RESPONSES_BASE_URL: Generic Responses API base URL; defaults to https://api.openai.com/v1; AGENT_WEB_SEARCH_MESSAGES_API_KEY: Generic Anthropic Messages API key for web search grounding; AGENT_WEB_SEARCH_MESSAGES_BASE_URL: Generic Anthropic Messages API base URL; defaults to https://api.anthropic.com; TAVILY_API_KEY: Tavily API key for agent-oriented search; BRAVE_SEARCH_API_KEY: Brave Search API key; PERPLEXITY_API_KEY: Perplexity API key for structured search; YDC_API_KEY: You.com API key; EXA_API_KEY: Optional Exa key; free MCP fallback without it; PARALLEL_API_KEY: Optional Parallel key; free MCP fallback without it; ZHIPU_WEB_SEARCH_API_KEY: Zhipu standalone Web Search API key; ZHIPU_CHAT_SEARCH_API_KEY: Zhipu Chat Completions native web search key.

Before connecting

Check the maintainer’s instructions for account access, authentication and the supported client configuration. Once connected, inspect the tools your client receives and try a small task that matches the server’s stated purpose. The registry description does not include a verified, complete tool list.

Project and source

View the publisher’s registry entry · Project repository · Project website. This profile uses version 0.8.0 from the registry. Registration identifies the publisher namespace; it does not establish that this is a first-party integration for every service it connects.

Browse related search & web scraping MCP servers for tools that cover the same kind of work.