Local-first code intelligence over MCP: BM25/FTS5 + AST graph, impact analysis, project memory and safe writes. Text-first by default; optional dense. Works with OpenCode and Zed.
Find any file on your machine instantly — a whole-machine, sub-millisecond file-search MCP for AI agents (NTFS MFT + USN, Windows)
Free local web search/extraction router for AI agents. Go CLI + MCP, BYOK/free-first routing, keyless DDGS/Scrapling fallback, setup writers and client guides.
MCP server for arXiv, implemented in Java with Quarkus
This Model Context Protocol (MCP) server enables LLMs like Claude to perform internet research using the Perplexity API. It provides real-time, up-to-date information with source citations.
a macOS CLI and MCP server that controls the Perplexity Desktop App via Accessibility automation, turning your existing Pro subscription into a programmable search backend for Claude, OpenCode, and an
152-source digital library MCP server. Natural-language search, full-text reading, and cited answers across academic papers, books, law, government records, security advisories, news, and developer do
One MCP server for every database. Drop your .env, run docker compose up, and any LLM — Claude, Codex, Gemini, Cursor, Windsurf — can read your PostgreSQL, MySQL, ClickHouse, MongoDB, Redis, SQLite, a
A self-hostable knowhow graph for people and agents: situations, goals, methods, outcomes and revisions with linked sources. Public read access; owner-only content API.
Open-source MCP server that gives AI coding agents secure database access via DocKit and SqlKit bridges — 70+ SQL & NoSQL databases, local-first, read-safe by default.
Google Calendar CLI for humans and AI agents
Grok-powered web search, fetch and map in a single dependency-free Go binary — usable as a CLI or an MCP server over any OpenAI-compatible endpoint. 基于 Grok 的网页搜索、抓取与站点测绘,单文件零依赖 Go 二进制——既是 CLI 也是 MCP
Single-binary local RAG MCP server in Rust — MiniLM compiled in, hybrid BM25+dense, zero runtime downloads
Local AI conversation memory: projects as books, tasks as chapters, and cited evidence pages. Python + SQLite, adaptive pagination, MCP, and reproducible benchmarks.
An MCP server that provides web search capabilities with support for multiple search engines
MyWebSearch - multi-engine web search MCP server, CLI and local daemon. No API keys required.
This read-only MCP Server allows you to connect to Bing Ads data from Claude Desktop through CData JDBC Drivers. For full CRUD support, check out the first managed MCP platform: CData Connect AI (http
Model Context Protocol server for OpenSERP search engine API
Free unlimited Google web search MCP server for AI agents. Uses real Chrome via Kimi WebBridge — no API keys, no rate limits. 8 IDE/CLI support, one-click setup.
Persistent identity and memory across AI tools - mcp-native, local-first, framework-agnostic, production-ready.
Cognitive digital twins and behavioral modeling for AI agents.
MCP server for web search and page extraction across the free tiers of several providers
Google Search Console MCP: search analytics, URL inspection, site management.
MCP server exposing Google Search grounding via Gemini
arXiv search MCP server in Rust
MCP server written in Rust, providing LLM agents with real-time web search and information retrieval capabilities.
MCP server that connects AI agents to the Talordata SERP API for real-time structured search data.
Rust SDK and MCP server for Kagi search and summarization in apps and AI agents.
极易部署 • 高性能 • 低内存占用 • 云原生支持 •Java版本的Elasticsearch MCP服务
Shared messaging, store, search and MCP layer behind the WireCat CLIs
学术文献 Agent:检索 → 筛选 → 下载(校园网 / CARSI / WebVPN)→ Zotero 归档 | 21 MCP tools for Claude, Cursor and Codex
Rust CLI + MCP server — parallel multi-source search across 56 APIs (academic, news, finance, security, geo) with JSON output for Claude Code
Privacy-focused web search via SearXNG instances for Zed AI assistant
MCP server for symbolic (LSP) code search
Go port of myscrape — web-research MCP server (web_search/web_fetch/web_research) as a compiled binary.
完整目錄
網頁、文件與內部搜尋。共 1,885 個伺服器,最常用的排在前面。每個都會開啟發布者的頁面。