AI & LLM MCP servers let your assistant do more than generate text. They connect it to memory systems, databases, orchestration layers, forecasting engines, and code intelligence tools so it can carry context across sessions, inspect structured data, coordinate multi-step agents, and answer questions grounded in your actual systems. Instead of pasting schema snippets, repo files, or prior chat history into every conversation, you install a server that exposes those capabilities directly through MCP.
This category is most useful for developers building agentic workflows, internal AI tools, coding assistants, research automations, and data-heavy applications. If you need an assistant to remember decisions, query production-safe data sources, reason over large codebases, run specialized sub-agents, or work with domain-specific context like trade history or graph memory, these servers cover that layer. The main differences are where the context lives, how tools are exposed, how much autonomy the agent gets, and whether the server is optimized for local knowledge, enterprise databases, code search, or long-running multi-step execution.
“Coordinate a planner, coder, and reviewer to implement OAuth token refresh across our API service and open a PR summary with claude-flow.”
viaclaude-flow“Inspect the customer and invoice tables, then write the SQL needed to find unpaid enterprise accounts with genai-toolbox.”
viaMCP Toolbox for Databases“Connect to our staging Postgres instance, compare the schema to production, and list breaking changes with mcp-toolbox.”
viaMCP Toolbox for Databases“Research competitors' pricing pages, summarize the differences, and draft an action plan for our launch with praisonai.”
viaPraisonAI| # | Server | Stars |
|---|---|---|
| 1 | Claude Flow Ruflo is an AI orchestration platform that enables the deployment and coordination of specialized AI agents for complex software engineering tasks. It features self-learning capabilities and enterprise-grade security. | 64.3k |
| 2 | Codebase Memory Mcp codebase-memory-mcp is a code intelligence engine that efficiently indexes code repositories and answers structural queries. It supports 66 programming languages and operates as a single static binary across multiple platforms. |
An MCP server exposes tools and resources in a format AI assistants can call directly during a conversation. That means the assistant can inspect available actions, choose the right one, and use the results as context without you manually wiring each API call into the chat.
Usually no. Most of them provide memory, orchestration, database access, retrieval, or specialized computation around an existing model client such as Claude Code, Cursor, Claude Desktop, or ChatGPT.
Use a memory-focused server when you want the assistant to retain notes, decisions, entities, or prior interactions across sessions. Use a database-focused server when the assistant needs live access to structured application data and schemas.
No. They are also useful for normal interactive development work like exploring a schema, searching a codebase, storing project decisions, or generating forecasts from local data. The assistant becomes more useful even when every step is still user-directed.
1,236 servers
ruvnet
AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
DeusData
Persistent codebase knowledge graph. Survives session restarts and context compaction.
oraios
A powerful toolkit for coding, providing semantic retrieval and editing capabilities.
googleapis
MCP Toolbox for Databases enables your agent to connect to your database.
googleapis
MCP Toolbox for Databases enables your agent to connect to your database.
MervinPraison
AI Agents Framework with Self Reflection and MCP support
basicmachines-co
Local-first knowledge management with bi-directional LLM sync via Markdown files.
bergside
Design systems, UI prompts, and layout variations for AI coding tools.
winedarksea
Automated time series forecasting with model search, anomaly detection, and event risk analysis
mnemox-ai
MCP memory for AI trading agents. Store trades, recall similar setups, track strategy performance.
vkhanhqui
Figma MCP server with full read/write access via plugin — no API token, no rate limits. 73 tools.
neo4j-contrib
MCP Neo4j Knowledge Graph Memory Server
ai.smithery
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to control Unreal E…
samvallad33
Local-first cognitive memory server for AI agents with SQLite, smart ingest, and portable sync.
ggozad
Opinionated agentic RAG powered by LanceDB, Pydantic AI, and Docling
PV-Bhat
Metacognitive AI agent oversight: adaptive CPI interrupts for alignment, reflection and safety
ooples
Intelligent token optimization achieving 95%+ reduction through caching, compression, and 80+ tools
flytohub
Execution engine for AI agents. 412 modules: browser, file, Docker, data, crypto.
tomyud1
MCP server for Godot game engine integration — control the Godot editor with AI
zhongweili
An MCP server that provides image generation and editing capabilities
vectifyai
Reasoning-based RAG system for chatting with long PDFs. Supports local and online files.
ruvnet
Neural network swarm orchestration with WebAssembly acceleration and MCP integration
artokun
MCP server + Claude Code plugin for ComfyUI: run workflows, generate images, manage models & VRAM.
shinpr
Easy-to-setup local RAG server with minimal configuration
tugcantopaloglu
MCP server for full Godot 4.x engine control with 165 tools for AI-driven game development
vfarcic
AI-powered development platform for Kubernetes deployments and intelligent automation
us
Open-source web scraper for AI agents with scrape, crawl, and map tools
aayoawoyemi
Persistent memory infrastructure for AI agents. Identity, knowledge graph, and search.
Lyellr88
Universal MCP Server with advanced AI memory capabilities and semantic search.
smart-mcp-proxy
Local-first MCP proxy with BM25 tool discovery, security scanning, quarantine & ~99% token savings
codespar
MCP server for Sift — global ML fraud detection: real-time risk scoring, decisions, event ingestion
sachitrafa
Ebbinghaus-based persistent memory for Claude. Memories decay with time, strengthen on recall.
sachitrafa
Persistent memory for AI agents. Memories decay with time and strengthen on recall.
sonpiaz
Hand any social video to your AI agent — frames + transcript bundled. MCP server for watch-cli.
openaccountants
Open-source AI accounting skills verified by licensed accountants (tax, VAT, payroll).
openaccountants
Open-source tax skills for AI across 134 countries.
l33tdawg
Persistent, consensus-validated institutional memory for AI agents. Runs locally.
billy-enrizky
AI browser automation. Write async Python to navigate, click, type, and extract data.
varun29ankuS
Cognitive memory for AI agents — semantic search, Hebbian learning, knowledge graphs.
nhadaututtheky
Persistent memory for AI agents — 55 MCP tools, spreading activation recall.
CrowdStrike
Connects AI agents with CrowdStrike Falcon for security analysis and automation.
RLabs-Inc
Gemini 3 MCP server with 30+ tools: images, video, research, TTS, code exec & CLI
“Store the architecture decisions from this chat as markdown notes and recall them when I ask why we chose Kafka with basic-memory.”
viaBasic Memory“Index this monorepo and show every place where the billing service constructs SQL queries with codebase-memory-mcp.”
viaCodebase Memory“Generate a 12-week demand forecast from our weekly sales dataframe and flag the series with unstable seasonality using autots.”
viaAutots| 3 | Serena Serena is an MCP server that gives coding agents semantic code retrieval, editing, refactoring, debugging, and memory tools at the symbol level. It connects to clients via MCP and uses either language servers or a JetBrains IDE backend for code analysis. | 26.4k |
| 4 | Mcp Toolbox MCP Toolbox for Databases is an open source Model Context Protocol (MCP) server that connects AI agents, IDEs, and applications directly to enterprise databases. | 16.0k |
| 5 | Genai Toolbox MCP Toolbox for Databases is an open source Model Context Protocol (MCP) server that connects AI agents, IDEs, and applications directly to enterprise databases. It provides prebuilt tools for instant data exploration and a framework for building custom tools. | 16.0k |
| 6 | Praisonai An MCP server that exposes PraisonAI agents and a large set of related tools to MCP clients such as Claude Desktop, Cursor, VS Code, and Windsurf. It supports agent execution, workflow orchestration, web search, crawling/scraping, file operations, code execution, and other utility functions. | 8.4k |
| 7 | Basic Memory Basic Memory allows users to build a persistent knowledge base through natural conversations with Large Language Models (LLMs), utilizing Markdown files stored locally. It employs the Model Context Protocol (MCP) for integration with compatible LLMs. | 3.4k |
| 8 | Typeui TypeUI provides an MCP server and design resources that help AI coding tools generate better UI by supplying design systems, prompts, and layout variation guidance. The repository links to installation guides for many IDEs and AI assistants that connect to the hosted TypeUI MCP server. | 1.5k |
| 9 | Autots AutoTS is a Python package for time series forecasting that enables rapid deployment of high-accuracy forecasts at scale, utilizing various forecasting models and transformations directly on Pandas DataFrames. | 1.4k |
| 10 | Tradememory Protocol TradeMemory is a memory layer for AI trading agents that records and recalls past trades and outcomes, ensuring compliance with regulatory requirements for algorithmic decision documentation. | 1.4k |