Deep Web Research Server

MCPOpen SourceMIT53.9

by qpd-v • Search & Retrieval

An MCP server that performs intelligent web research and content extraction for Claude, integrating Google search, page visiting, and analysis tools.

Example Use Cases

1
Gather up-to-date web evidence and citations to support answers or claims during a Claude session.
2
Extract and convert webpage content into clean Markdown for downstream summarization, analysis, or citation.
3
Run parallel, rate-limited Google searches with deduplication and relevance scoring to discover high-quality sources quickly.

Description

This Model Context Protocol (MCP) server enables agents to perform real-time web research by running batched Google searches, visiting pages, extracting and cleaning content, and scoring relevance with TF‑IDF and readability metrics. It includes tools for parallel search queuing, deduplication, structured data extraction, and a visit_page tool that returns Markdown content. The server is designed to work within MCP timeout constraints and integrates with Claude Desktop for seamless use.

Quick Actions

View on GitHub

Security

Scanned 2 month(s) ago

Risk Level

MINIMAL

Read-only data retrieval, no side effects

Trust Score

D44/100

4/17 checks passed

Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.

Quick Stats

Service TypeMCP

Pricing ModelFree

Capabilities0 Tools / 0 Prompts / 0 Resources

Ownerqpd-v

CategorySearch & Retrieval

DependenciesStandalone