Best MCP AI Skills & MCP Servers

Nookplot MCP server — connect any MCP-compatible agent to the Nookplot network

Stdio relay and fallback MCP server for Agoragentic's hosted Triptych OS (Agent OS) Router / Marketplace. Exposes registration, search, provider preview, paid execution, and receipt-backed status tools.

Local MCP orchestrator for managing Codex, Claude, and Cursor worker runs through a persistent daemon, with a daemon-owned orchestration_run engine for multi-step workflows.

MCP server for spawning Claude Code agents in cloned repos — branch your agents

MCP server for atrib. Synthesizes a narrative across N records via an OpenAI-compatible LLM so agents read context, not raw record bytes.

Local implementation of agent-orchestrator MCP server

Project-local RAG memory MCP server — knowledge graph + multilingual vector + FTS5 in a single SQLite file. Per-project isolation, 30 MCP tools, codepoint-safe chunking (Korean/CJK/emoji).

List directory contents — entries, tree, glob, find.

MCP server for searching and getting up-to-date information about NPM, Cargo and NuGet packages.

Studio Signal MCP Server — AI-powered M&E intelligence via Model Context Protocol

Model Context Protocol server exposing the b402 Solana SDK as agent-callable tools (Claude Code, Cursor, any MCP runtime)

Project-level understanding without reading all files — detect framework, language, conventions, and architecture.

MCP server for pūrmemo - AI conversation memory that works everywhere. Save and recall conversations across Claude Desktop, Cursor, and other MCP-compatible platforms. Intelligent context extraction, smart titles, living documents.

MCP server that gives AI agents instant access to decompiled Java code from Maven dependencies. Blazing fast Java class analysis for Claude, Cursor, and other LLM-powered IDEs.

Multi-AI Rules MCP Server - One source of truth for AI coding rules across all AI assistants

A local context graph for AI coding agents.

Production-grade AI coding toolkit — 70 agents (incl. devil-mode adversarial crew), 194 skills, 97 commands, parallel multi-agent commands, semantic intent routing, self-learning memory, and a built-in MCP server (16 tools / 6 prompts / 377 resources) tha

Model Context Protocol (MCP) server for Zephyr Scale test case management with comprehensive STEP_BY_STEP, PLAIN_TEXT, and BDD support

TypeScript declarative UI component library for MCP apps. Wire-compatible with Python prefab-ui.

Full-text search with TF-IDF ranking — index files, search with relevance scoring, suggest completions.

Format data into various output formats — pretty JSON, YAML, markdown, ASCII tables.

Generic config-driven MCP wrapper for atrib. Spawns any upstream MCP server and applies @atrib/mcp middleware so every tool call is signed and logged. Wrap any MCP via JSON config, no per-server code.

MCP server and client for Agentick — standalone, pluggable, world-class

Local-first nutrition MCP server for AI agents: food search, barcode lookup, meal photos, hydration, goals and safe intake logging.