Aas-ee/open-webSearch

Health v2 · maintenance only

Adoption not part of the score

1747 stars · 185 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Open-WebSearch is a TypeScript tool providing an MCP server, CLI, and local daemon for multi-engine web search and content retrieval without requiring API keys. It aggregates results from engines like Bing, DuckDuckGo, Brave, and Baidu, and can fetch article content from pages like CSDN and GitHub READMEs.

Use cases

• search the web from an AI agent without API keys
• add web search to Claude Desktop or Cursor via MCP
• scrape search results from Bing or DuckDuckGo in scripts
• fetch article content from URLs for LLM context
• run a local HTTP search service for automation
• give coding agents live web retrieval capabilities

When to choose

• you need keyless web search integrated into MCP clients or agent workflows
• you want a CLI or local daemon for scripted search and page fetching
• you need multi-engine coverage including Chinese sources like Baidu and CSDN

When to avoid

• you need a compliant official search API with guaranteed terms of service
• you require high-volume production scraping with rate-limit guarantees
• you need advanced search features like semantic ranking or filters

Facets

service · maturity active

search-engine web-scraping mcp cli http-server developer-tools artificial-intelligence cli cross-platform self-hosted mcp-server no-api-keys multi-engine-search web-content-fetching local-daemon agent-tools search ai-agents command-line nodejs

Member repositories

For agents

markdown · JSON · MCP: product_card(name="Aas-ee/open-webSearch")