Claude Code plugin — structural guardrails for multi-agent software development. RTM-based evidence, cross-model adversarial audit, 7 MCP tools, worktree isolation, HITL retrospective. DOI: 10.5281/zenodo.19108370

Dynamic MCP integration for AI agents — search 1,960+ servers, install with HITL approval

StewardFlow 是一个基于 FastAPI 的 ReAct + HITL（人机协作）智能体系统，没有依赖第三方框架和中间件，纯本地运行的解决方案。提供可视化前端工作台、可追踪的执行日志，以及可扩展的工具与 MCP 服务接入。它适合快速构建“可控、可追溯、可复现”的智能助手。

A HITL (Human-In-The-Loop) MCP server that runs on CLI.

Enabling Claude Desktop to access Cogito HITL Service from AgentMP

A secure Layer 7 firewall and proxy for the Model Context Protocol (MCP) that intercepts `stdio` traffic, enforces security policies, and supports Human-In-The-Loop (HITL) approval workflows to protect local resources.

Compliance authoring for OSCAL artifacts based on MCP Server and GenAI agentic solution with HITL

HITL over MCP

A production-ready, state-persistent MCP server for Human-in-the-Loop (HITL) interaction. Enables AI agents to suspend execution, ask structured questions to users, and resume contextually across sessions — friendly with ACP.

Production-ready skill library for building LangGraph AI agents — RAG, multi-agent, HITL, memory, cost optimization, and MCP integration.

Autonomous SRE & Cloud FinOps Orchestrator: LangGraph + Gemini + Qdrant + MCP to detect underutilized AWS resources and open Terraform PRs with HITL approval

Personal AI Employee — Silver Tier: Local-first autonomous AI agent with Gmail/WhatsApp/LinkedIn watchers, MCP servers, HITL approval, and web dashboard. Built with Claude Code + Obsidian.

Autonomous AI Employee: monitors Gmail, WhatsApp & finances using Groq LLM. Bronze→Silver→Gold→Platinum tiers with Obsidian vault, HITL safeguards, MCP servers & Oracle VM cloud support. 100% free

Agentic PE intelligence platform for the Forbes AI 50: Airflow-orchestrated ETL + RAG + structured extraction, wrapped with MCP tools, a ReAct-based supervisor agent, HITL review, and Dockerized FastAPI/Streamlit dashboards.

Created a ChatGPT clone using advanced langGraph concepts such as memory, persistence, HITL (Human In The Loop), streaming, etc. Implemented features like chat resumption, MCP tool integration, RAG, and observability using LangSmith.

A production-ready Agentic Procurement App built with Google ADK (Agent Development Kit) and Python. Features autonomous agents, HITL workflows, and MCP integration with custom UI. The ultimate ADK Python blueprint for enterprise AI.

A production-grade Autonomous SRE Agent for self-healing infrastructure. Built with LangGraph (Orchestration), Groq LPU (Inference), Model Context Protocol (MCP) for tool isolation, and NeMo Guardrails for deterministic safety. Features Human-in-the-Loop (HITL) approval via Streamlit.

🚀 Silver Tier (Vector): An autonomous Personal AI Employee built with Claude Code, Agent Skills, and MCP. Features 24/7 file/email monitoring, web research, LinkedIn drafting, multi-step planning, and strict Human-in-the-Loop (HITL) safety protocols.

🛰️ Vanguard Agent is a Security-as-a-Service Multi-Agent Orchestrator for complex, autonomous, and governed workflows. Built with Next.js 16 and LangGraph, it goes beyond RAG with HITL-gated ReAct loops, persistent memory, and an MCP-integrated supervisor-worker model for mission-critical multi-step execution with auditor-grade traceability.

HITL - MCP Implement

AidaAgent HITL MCP Server

A typescript reference implementation of enforced HITL for MCP tools using two approaches: OTP and WebAuthn (ed25519 signatures)

Human-in-the-Loop MCP Server for WeCom (企业微信)

Universal MCP proxy with Human-in-the-Loop approval flow via Telegram

MCP Server Code

MCP service with Embabel and human-in-the-loop

No description available

Human-in-the-Loop refund processing using MCP

An HITL MCP service that supports approval, notification, and information collection.

A simple but powerful agentic harness with MCP integration and HITL features