Aas-ee/open-webSearch

Overview

A TypeScript-based MCP server, CLI, and daemon that enables agents to perform multi-engine web searches and content retrieval without requiring API keys. It uses skill-guided workflows to orchestrate searches across multiple engines and parse results locally.

Best for

Best for
Developers building autonomous agents that need flexible, API-key-free web search integration

Use cases

• Letting autonomous agents search the web and retrieve page content
• Running local search pipelines without external API dependencies
• Integrating multi-engine search into MCP-compatible development workflows

How to use

Install

npx skills add https://github.com/Aas-ee/open-webSearch --skill open-websearch

Tools exposed

• ENABLE_CORS
• CORS_ORIGIN
• DEFAULT_SEARCH_ENGINE
• USE_PROXY
• PROXY_URL
• FAKE_IP_CIDRS
• FETCH_WEB_INSECURE_TLS
• ALLOWED_SEARCH_ENGINES
• SEARCH_MODE
• PLAYWRIGHT_PACKAGE
• PLAYWRIGHT_MODULE_PATH
• PLAYWRIGHT_EXECUTABLE_PATH
• PLAYWRIGHT_WS_ENDPOINT
• PLAYWRIGHT_CDP_ENDPOINT
• PLAYWRIGHT_HEADLESS
• PLAYWRIGHT_NAVIGATION_TIMEOUT_MS
• OPEN_WEBSEARCH_PROFILE_DIR
• MCP_TOOL_SEARCH_NAME
• MCP_TOOL_FETCH_LINUXDO_NAME
• MCP_TOOL_FETCH_CSDN_NAME

Tested with

Claude Desktop, Cursor, Continue, VS Code

Notes

A TypeScript-based MCP server, CLI, and daemon that enables agents to perform multi-engine web searches and content retrieval without requiring API keys. It uses skill-guided workflows to orchestrate searches across multiple engines and parse results locally.

1,338 stars on GitHub. Last updated 2026-05-16. Licensed Apache-2.0.

Use cases

• Letting autonomous agents search the web and retrieve page content
• Running local search pipelines without external API dependencies
• Integrating multi-engine search into MCP-compatible development workflows

Pros

• No API keys needed, reducing setup friction and costs
• Supports multiple search engines for diverse retrieval strategies
• Modular skill-guided workflow design for adaptable agent behavior

Cons

• Limited to web search and content retrieval, not a general-purpose MCP server
• Requires running a daemon or CLI process, adding operational overhead
• May depend on public search engines that could throttle or block automated access

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.