WebSearch MCP Server — Claude Code MCP server
Runs multi-engine internet and academic searches without any API keys. Parses PDF files and safely retrieves external web pages for the agent.
Open on GitHubWhat WebSearch MCP Server does
WebSearch MCP Server connects an assistant to multiple search backends. It handles standard web queries alongside scholarly literature lookups. When an agent opens an external link, the built-in fetcher loads the page safely.
The server also extracts text directly from PDF documents. No paid API tokens are required. Searches run out of the box without external account credentials.
Authored by daidaiJ under the MIT license, the tool runs with Claude Code, Codex, Cursor, GitHub Copilot, Windsurf, Cline, and Zed.
What it helps with
- run web queries without registering paid search tokens
- find academic papers and research publications
- pull readable text from PDF files into the prompt
- fetch external web pages securely during conversations
Example prompts for the agent
- Search academic sources for recent papers on transformer latency
- Fetch this documentation link and extract its key points
- Read
report.pdfand summarize the methodology section
How to install WebSearch MCP Server
Install
The MCP registry has no launch command for it: see the repository README
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about WebSearch MCP Server
How to install WebSearch MCP Server in Claude Code?
Add the server to Claude Code using the daidaiJ/websearch-mcpserver repository in your MCP client setup.
What does WebSearch MCP Server do?
It handles web and scholarly searches across multiple engines, reads PDF files, and securely fetches remote web pages.
Does WebSearch MCP Server require an API key?
No API key is needed. The server executes search queries without external account credentials.
