human-in-the-loop
32 servers · 1,237★ total
The action firewall for AI agents. Enforce policy and human approval before risky tool calls, shell commands, workflows, and production changes, with auditable evidence.
Harness-oriented agent system framework for production-grade LLM agent applications
Agent Identity Protocol - Zero-trust security layer for AI agents. Policy enforcement proxy for MCP with Human-in-the-Loop approval, DLP scanning, and audit logging.
Save your Cursor monthly quota! Unlimited AI interactions in one conversation via MCP feedback loop.
Achieve infinite conversation turns in a single API request via turn-mcp. Self-hosted MCP server with browser console for human-in-the-loop AI agents.
The agent harness you steer — structured context, automatic staleness gates, and block-by-block review, all bound to your code. Local-first, MCP-native, open source.
The open-source safety layer for AI agents — block unsafe tool calls, require approval, enforce budgets, audit, replay.
Bring your teammates into your AI coding agent's planning loop. An MCP server that routes your agent's design questions to a shared web page where teammates discuss in their browser — no install, no signup. Works with Claude Code, Codex, OpenCode & Gemini CLI.
Open-source browser automation broker for AI agents: persistent Chrome profiles, proxy-aware identities, remote API, MCP tools, browser-use/OpenBrowser adapters, human auth handoff, telemetry, and audits.
Auditable MCP + cross-agent harness for autonomous and human-guided academic writing—from verified evidence to reviewed DOCX/PDF.
The agent that learns how you decide - and proves it by never deciding for you. A local-first, loop-native runtime for a team of AI agents on your machine.
Markdown-first multi-model AI orchestration in one tmux session. A single coordinator (Chrono) routes scoped markdown task packets across 4 frontier model lanes (GPT-5.6 Sol / Claude Opus 5 + Fable 5 / Gemini 3.5 / Kimi K3) and 73 role-based specialists — with cross-family review gates, git-snapshot safety rails, and no server. Dogfooded.
MCP-powered memory, policy, and experience layer for safer AI agents.
End-to-end encrypted inbox for AI agents — deliver artifacts to humans via HTTP API.
GASADK: Agent Development Kit for Google Apps Script. Build AI agents with Gemini API, MCP, A2A, Agent Skills, Human-in-the-Loop (HITL) suspension/resumption, token quota safeguards, and dynamic Google API MCP servers.
LiAgent OS is a local-first AI agent OS for building a private personal assistant with governed autonomy across local models and hybrid cloud services. It brings conversation, tool use, multi-agent orchestration, task scheduling, heartbeat execution, human approval, and auditability into one long-lived loop.
MCP server for human-in-the-loop surveys, A/B preference tests, ratings, and rankings. Get real human feedback inside Claude Code, Claude Desktop, Cursor, Windsurf, and any MCP client — powered by Datapoint AI.
Local-first permission gateway for AI agents: connector-based SSH, Postgres, and Redis access with scoped tokens, human approval, audit logs, and encrypted local credentials.
面向个人漫画创作者的开源漫画工作台，支持故事与资产管理、页面与画格编排、预览与导出，并通过 MCP + Skill 接入外部 AI Agent。
A QuickBooks MCP server built for real books. Exposes QuickBooks Online as callable tools over the Model Context Protocol, with every create, update and delete gated behind human approval and a plain-English summary, and all QuickBooks text treated as untrusted input rather than instructions.
Render agent-written Markdown as a live local page, and send the human's click back to the agent as typed data. Human-in-the-loop approval gates for coding agents.
Zero-dependency TypeScript framework for production AI agents: durable execution, long-term memory, hybrid RAG, MCP tool calling, human-in-the-loop approval, planning and CodeAct sandboxes. One streaming API for Claude, GPT, Gemini, Grok, Mistral and DeepSeek — Node, Bun, Deno, serverless and edge.
Make every AI chart write a reviewable proposal. Draft protocol for agent-initiated FHIR writes (Proposal → Decision → Commit → Audit) with web + MCP implementations and a conformance suite. Runs over Medplum, HAPI, Firely, Aidbox, or Oystehr.
Community-built Codex–COMSOL bridge using officially supported MATLAB MCP and LiveLink integration layers, with native multiphysics API access, task-level execution, same-model Desktop observation, and verified human takeover.
Discord x Claude Code bridge via MCP (Model Context Protocol). Human-in-the-loop AI assistant integration with forum threads, real-time communication, and AI-powered title generation.
The guardian layer for AI agents — identity, secrets, audit via MCP. Gate + Vault + Watch.
MCP server that saves your Claude Code, Cursor, and Codex sessions as versioned Markdown reports — with human-in-the-loop review and Apple Pencil / S Pen annotation.
Runtime governance for AI agents — 57 MCP tools, embedded engine, no key required. Decision gates, forensic audit, EU AI Act / NIST / SOC 2 mapping. Model-agnostic.
The natural-language command layer for an AI workforce — direct governed AI agents to deploy cloud infra, train ML, and run research, all under human policy & approval. Agentic command-and-control.
Small MIT-licensed runtime for bounded LLM traversal, inspired by FPF and focused on practical agent behavior.
MCP server for market data, SEC filings, macro and risk tooling across Webull, Saxo and IBKR — with order execution behind a human approval step the assistant cannot reach. 41 tools, filtered to what your broker can serve.