Web & Search

Paper Search is an open-source Model Context Protocol (MCP) server designed to give AI agents direct access to global academic repositories. It enables researchers to search, retrieve metadata, and extract full-text content from sources like arXiv and PubMed through a unified interface. Its key differentiator is the support for eight simultaneous academic platforms, streamlining the transition from discovery to deep analysis within LLM workflows.

ai.com.mcp/skills-search is a discovery tool for the Model Context Protocol (MCP) that allows AI agents to search and retrieve procedural knowledge from the skills.sh registry. It acts as a bridge between an agent's connectivity (MCP) and its instructional guidance (Agent Skills), enabling 'on-the-fly' capability expansion.

Linkup is an API-first search engine specifically engineered to provide AI agents and RAG pipelines with high-accuracy, real-time web context. It differentiates itself through superior factual grounding, outperforming major competitors like Perplexity on benchmarks such as SimpleQA, and offering developer-centric features like structured JSON outputs and sub-second latency.

Google Scholar Search (MCP) is an open-source Model Context Protocol server that bridges AI models like Claude, Gemini, and Cursor with academic data. It allows AI agents to perform real-time scholarly searches and retrieve paper metadata through a standardized, streamable interface. It is primarily designed for researchers and developers building autonomous AI research assistants.

YouTube-MCP is a Model Context Protocol server that allows AI agents like Claude to interact directly with YouTube content, including searching videos and fetching deep metadata. It is primarily for researchers and developers who need to perform automated video analysis, summarization, and RAG (Retrieval-Augmented Generation) without manual copy-pasting. Its key differentiator is the ability to retrieve timestamped, multilingual transcripts natively within an AI's context window.

Context7 is a specialized retrieval-augmented generation (RAG) tool designed to provide AI coding assistants with up-to-date, version-specific documentation. It eliminates LLM hallucinations by fetching real-time code snippets and API references directly from official sources into prompts via the Model Context Protocol (MCP).

Exa Websets is an AI-powered entity discovery and list-building tool that uses neural search to find and verify companies, people, or research papers from across the live web. Unlike static databases, it allows users to define custom verification criteria and enrichment fields in natural language, making it ideal for GTM teams and researchers targeting niche or fast-moving markets.

Grokipedia-MCP is a Model Context Protocol (MCP) server that enables AI agents to programmatically search, retrieve, and analyze articles from xAI's Grokipedia encyclopedia. It provides specialized tools for full-text retrieval, citation extraction, and guided research workflows, acting as a bridge between LLMs like Claude and xAI’s AI-curated knowledge base.

NGSS Standards Explorer is a specialized Model Context Protocol (MCP) server that provides AI agents with structured access to the Next Generation Science Standards. It is designed for educators and developers to help AI models navigate complex three-dimensional science frameworks (SEPs, DCIs, and CCCs) for curriculum planning and standards alignment.

Google Drive is a industry-leading cloud storage and collaboration platform that serves as the backbone of the Google Workspace ecosystem. It allows users to store, share, and co-edit files in real-time, differentiating itself through seamless integration with Google's suite of productivity tools and powerful AI-driven search capabilities.

Brave Search API provides access to a fully independent, privacy-centric web index that is not dependent on Google or Bing. It is specifically designed for developers and AI researchers who need structured web data, local results, and optimized context for LLM grounding and RAG pipelines. Its key differentiator is the 'LLM Context' endpoint, which delivers pre-extracted, token-efficient web content directly to AI agents.

A Model Context Protocol (MCP) server that connects AI agents to Semantic Scholar and arXiv for real-time academic research. It is designed for researchers and developers who need LLMs to perform grounded literature searches, citation analysis, and full-text PDF extraction. Its key differentiator is the ability to fetch full-text content from arXiv and Wiley open-access sources without requiring an upfront API key.

This MCP server enables AI assistants to interact directly with Obsidian vaults stored on GitHub, allowing for seamless natural language querying of personal notes. It is designed for power users who want to use their private knowledge base as context for LLMs without manual data entry. Its key differentiator is the use of the GitHub API, providing a bridge to remote vaults that local-only file connectors cannot reach.

RSS Reader by kwp-lab is a Model Context Protocol (MCP) server that enables AI agents, specifically Anthropic's Claude, to subscribe to and read RSS feeds. It allows LLMs to fetch live news entries and extract clean article content in Markdown format for real-time analysis. It is primarily for developers building agentic workflows that require automated information monitoring.

ai.fiber/mcp is an open-source Model Context Protocol server that enables AI agents like Claude to perform real-time B2B lead enrichment and contact searches directly within a chat interface. It is designed for sales and marketing professionals who want to automate prospecting without switching between browser tabs, offering a high-efficiency bridge between business intelligence and generative AI.

Exa is a neural search engine designed specifically for LLMs and AI agents, using vector embeddings rather than keywords to find high-signal web content. It serves as a data layer for developers, providing clean, structured data and specialized tools like 'Exa-code' to ground agents in real-time documentation and code examples.

Notion is a modular "all-in-one" workspace that integrates notes, databases, wikis, and project management into a single, block-based platform. It is designed for individuals and teams who require a highly customizable "second brain" or a centralized source of truth. Its key differentiator is its relational database system, which allows users to view the same information across multiple layouts like boards, calendars, and galleries.

Docfork is a Model Context Protocol (MCP) server designed to feed up-to-date, AI-optimized documentation from over 10,000 libraries directly into AI agents and code editors. It eliminates AI hallucinations caused by stale training data by providing a real-time retrieval layer for technical documentation.

DuckDuckGo & Felo AI Search is an open-source Model Context Protocol (MCP) server that provides AI assistants with real-time web search and content extraction. It is designed for developers building AI agents who need a privacy-focused, zero-cost alternative to paid search APIs. Its key differentiator is the ability to provide AI-powered answers via Felo AI and web results via DuckDuckGo scraping without requiring any API keys.

Gmail is the industry-standard email platform by Google, providing comprehensive inbox management with AI-powered sorting, search, and security. It is designed for both personal users and enterprises, serving as the central hub of the Google Workspace ecosystem. Its key differentiator is its unmatched integration with productivity tools like Google Drive and Calendar, paired with robust spam filtering.