Servers
Browse enriched MCP servers by capability — filter on inferred permissions (all-of), transport, verified status, and tool count. Only servers with a captured capability surface appear here.
53AI Hub is an open-source AI portal, which enables you to quickly build a operational-level AI portal to launch and operate AI agents, prompts, and AI tools. It supports seamless integration with development platforms like Coze, Dify, FastGPT, RAGFlow.
MCP Security Solution for Agentic AI — real-time proxying, behavior analysis, and malicious tool detection
Drop-in replacement for better-sqlite3 using WebAssembly. Zero native modules — works on any Node.js version and platform. Includes FTS5.
Guidance Control Plane - Compiles, retrieves, enforces, and evolves guidance rules for Claude Code sessions
Self-Optimizing Neural Architecture (SONA) for Claude Flow — adaptive learning, trajectory tracking, pattern reuse, 7 RL algorithms (PPO/A2C/DQN/Q-Learning/SARSA/Decision Transformer/Curiosity), Flash Attention, MoE routing, LoRA, EWC++ for continual lear
FrontMCP UIpack - HTML shell builder, pluggable import resolver, and NPM component loader for MCP UI (React-free core)
Shared utility functions for FrontMCP - string manipulation, URI handling, path utilities, and more
A self-healing harness for AI coding agents. Gives GitHub Copilot (and Microsoft Scout) objective verification + bounded recovery so it stops shipping silently-broken work. Cut Copilot's silent-failure rate 40%->0% on 20 live sessions. Rust core, MCP + CLI, MIT.
Curated technical knowledge base: 785+ dense reference articles across 26 domains. Agent-first format with wiki-links, gotchas, runnable code. Not tutorials — verified references for AI agents and engineers.
44 plug-and-play skills for OpenClaw — self-modifying AI agent with cron scheduling, security guardrails, persistent memory, knowledge graphs, and MCP health monitoring. Your agent teaches itself new behaviors during conversation.
This project was done as Master's Thesis in Institute of Particle Technology within TU Braunschweig. An autonomous LLM agent built using the Model Context Protocol (MCP) that optimizes symbolic regression on real industrial datasets from particle technology research.
✨ A curated list of awesome GitHub instructions, prompt, skills, MCPs and agent markdown files for enhancing your GitHub Copilot AI experience.
CookHero是一个基于 LLM + RAG + Agent + 多模态的智能饮食与烹饪管理平台，支持智能菜谱查询、个性化饮食计划、AI 饮食记录、营养分析、Web 搜索增强，以及可扩展的 ReAct Agent / Subagent 工具体系，帮助厨房新手轻松成为“烹饪英雄”。
Local graph memory for AI coding agents, giving MCP-compatible tools precise codebase context before they edit.
OpenCode plugin that routes tool, MCP, and skill usage with lightweight task-aware guidance.
Your on-call's minute-15 incident page, at minute 1. A Google ADK multi-agent SRE assistant: when a Dynatrace problem opens, six agents run the pre-mortem an experienced SRE would and post a ranked causal brief to Slack + Grail in ~90s. Dynatrace MCP wired as an ADK McpToolset.