The enterprise-grade CrewAI alternative for Go. Orchestrate strictly-typed, parallel AI agents with LangGraph-style Flow persistence, distributed A2A protocols, Model Context Protocol (MCP), and Human-in-the-Loop training.

brain-inspired MCP for persistent efficient context across AI agents

AI-powered job hunting toolkit built on Claude Code: agents, MCP server, coaching frameworks, and automation scripts

Graph-based workflow enforcement and persistent memory for AI coding agents. Define structured workflows in YAML. Enforce them at tool boundaries via MCP. Build a persistent knowledge graph that grows with every query and knows when its sources have changed.

Persistent memory for AI agents — OpenClaw plugin + MCP server

SQLite-backed MCP Memory Server with WAL concurrent safety, FTS5 search, session tracking, task management, and cross-machine bridge sync

MCP server for Amazon Connect - +85 tools for AI assistants to manage contact centers

excel mcp server

A structured data reader and writer like 'jq' and 'yq' for AI Agents

MCP server to generate PPTX presentations from MD file

Yet Another SAP MCP Server

ZenStack Remote MCP Server with Authorization

MCP Server for MSSQL (Microsoft SQL Server)

Unity Editor MCP plugin — AI agents drive Unity via Model Context Protocol. Apache 2.0, ~200 domain-specialized tools including procedural generation, simulation, physics, rendering, and AI behavior from published research papers.

DevKit (TXC) CLI and MCP Server for automation of Power Platform developer workflows

sbox mcp documentation

OAuth 2.1 + OIDC bridge for private MCP servers. Stateless, replay-safe, audit-defensible. Bring any IdP — Keycloak to Entra.

Hermes Brain — Multi-agent orchestration for Hermes Agent. SQLite-backed coordination layer, MCP-compatible, zero-token coordination overhead.

Your AI setup. Everywhere you work. Syncs Claude Code agents, skills, plugins, and MCPs across machines.

Proof of concept project about LLM usage inside Ragnarok Online via sse MCP integrated to OpenKore

mcp-recall compresses MCP tool outputs (94 KB → 3.5 KB · 96%) and stores full results in SQLite for retrieval — up to 30x more tool calls per session for heavy MCP workloads.

MCP server for reading and editing Word (.docx) documents with track changes, comments, footnotes, and structural validation

MCP server for multi-repository code indexing with Qdrant vector search and LSP integration. Provides unified context across frontend, backend, and microservices for AI coding assistants like Claude Code.

mcp safe fetch