JSPM

Deployment tool and support utility for AI context. Copies agents, skills, commands, rules, and behaviors into the paths each AI platform reads (Claude Code, Codex, Copilot, Cursor, Warp, OpenClaw, and 6 more) so one source of truth works across 10 platfo

High-performance job queue for Bun & AI agents. SQLite persistence, cron scheduling, priorities, retries, DLQ, webhooks, native MCP server. Zero external dependencies.

The explainable agent framework — backtrack a wrong answer to the exact context that caused it (evidence, not guesses). Built on footprintjs.

Super Humans: one handoff_to_human tool for blocked AI agents using Browserbase, Daytona, and remote browsers.

The Sudo Command for AI Agents. Execution Security for Claude Code, Codex, Gemini, Cursor, Opencode, Pi, and any MCP server.

h2a — the unified CLI + core for human-to-agent coordination.

Question boards with rich blocks (markdown, charts, mermaid, tables, code, diffs, video, sandboxed HTML), clickable local file-links, and element-level annotations for AI coding agents (Claude Code, Codex, …): ask users structured questions, present inter

Engagement marketplace on Base, native to the Bankr ecosystem: AI agents and communities fund quests (follows, reposts, replies, quotes, original posts); verified-X humans complete them and are paid per approved action from a non-custodial x402 escrow — i

AI Agent Governance for TypeScript — policy enforcement, scoring, compliance, and audit for AI agents

Pane relay client: typed HTTP + WebSocket operations against a Pane relay. Framework-free.

Permission hooks for AI coding agents: route tool approvals through Pushary push notifications

Free MCP server for developers adding human review gates, prompt risk checks, FAQ review, and CRM note safety to AI tools.

MCP server for AI-orchestrated human QA testing. Let AI agents request real human testing of web applications.

AI-native web browser runtime for autonomous agents with human supervision

An open-source AI coding agent runtime and terminal CLI for real repositories, with persistent memory, sessions, approvals, heartbeat, and a browser control plane

Transport-neutral orchestration runtime for loom — the headless driver-loop, the directive→wire adapter, and the conformant delivery composition shared by every transport.

n8n community node for HumanStep — add human decision steps to your automation workflows

Generic finite-state-machine kernel for multi-agent LLM workflows — atomic state, replay-deterministic, zero runtime dependencies.

Realm workflow execution engine — state machine, evidence capture, and service adapters for agent-driven workflows.

Shared wire-response types for loom transports.

Command-line client for the Pane relay: create panes, inspect state, send and watch events.

Human-in-the-loop interaction tool. The `AskHuman` CLI pops up a window (or uses Telegram / Slack / DingTalk / Feishu) to collect a human response and return it to the AI — usable standalone or as a dependency.

Native Telegram approval, choice, and reply workflows for n8n without external response pages

Zero-config shuttle provider for loom — runs agents through the host with no API key.

A control plane for the human bottleneck.

Long-lived supervisor over loom's headless driver-loop — drives a project's task server-side, parks on a human gate and wakes on the answer, retries transient failures, recovers in-flight work on restart, and owns the worktree lifecycle.

OKed for the Claude Agent SDK. A ready-made PreToolUse hook that gates sensitive tool calls behind a human approval push to your phone.

loom's persistent control layer — the global config + secrets + project catalog, and the one layered resolver every face (CLI today, server/UI later) reads. A leaf: it depends on nothing else in loom and learns no domain. Configure keys/models once; inher

MemberJunction: Angular components for AI agent request management - panels and dialogs for viewing and responding to agent feedback requests

OKed plugin for OpenClaw. Gates skill and tool calls behind a human approval push to your phone.

One Cursor conversation, unlimited AI interactions - Save your monthly request quota! Interactive feedback loop for AI chat via MCP

Code review / implementation bundle for loom — agents, reference knowledge, schemas, and stack candidates.

Network control plane for loom — an HTTP transport over the headless driver-loop and a registry that supervises a fleet of projects from one process. Submit a task, read status, and answer a gate over localhost; intake adapters (a dashboard, a chat bot, a

Copy-paste MCP server: let your AI agent ask a human for approval on their phone. End-to-end encrypted, no accounts, no database.

MCP transport for loom — exposes the pipeline tools over stdio and assembles the runnable registry.

MCP server for Loopuman — The Human Layer for AI. Give AI agents instant access to verified human workers in 30+ countries. Post tasks, competitions, and social contests. Workers paid in seconds via blockchain.

Self-improving agent workflows: every run distills lessons into memory, and future runs retrieve them automatically. Cyclic graph engine with durable execution, per-node budgets, taint tracking, and human-in-the-loop gates.

The loom command-line installer — register the MCP server and authorize projects.

Model Context Protocol (stdio) server for Pane: lets any MCP client (Claude Desktop, Cursor, …) hand a human a rich interactive UI by URL and get structured data back.

Install loom in one step: npm i -g @loomfsm/pipeline && loom setup. Pulls the CLI, the MCP server, the default bundle, and the zero-config provider.

loom's web control plane — a React SPA over the @loomfsm/server HTTP API. A PEER client of the same resolver the CLI drives: edit config/secrets/models, browse projects, submit tasks, answer gates. Ships as PREBUILT static assets the server serves at / —

Parley core: the transport-agnostic seam, normalized Message, cursor/dedup engine, and dual-role MCP channel server. Zero backend dependencies.

Multi-model routing provider for loom — runs agents through OpenRouter's OpenAI-compatible API.

Human-in-the-loop approval workflows for AI agents

Production payment infrastructure for AI agents — real Stripe/Paystack/Lightning rails, autonomous shopping with escrow, Agent Credit Score (300-850), Ed25519 identity, behavioral finance, Merkle integrity, anomaly detection, human-in-the-loop approval. B

Typed async workflows with automatic error inference. Build type-safe workflows with Result types, step caching, resume state, and human-in-the-loop support.

Manage Cursor from your phone over Telegram: an MCP server + auto-spawned local worker that notifies you, asks you questions, and (optionally) runs headless Cursor agents you text it. Local, bring-your-own-bot, runs entirely on your machine.

Hand control from an agent to a human via a local web runbook.

AI-native Agent Runtime for HazelJS - stateful, long-running agents with tools, memory, and human-in-the-loop

Give the coding agent you already run (Claude Code, Codex, Gemini CLI, OpenCode, pi) an evolving body of knowledge, memory, actions, and guardrails — plain markdown notes it queries, runs, and grows from your real sessions, every change human-gated. Zero

The authorization layer for AI agents — allow/deny/ask policy on every tool call, human approvals, and tamper-evident receipts. Framework-neutral, zero dependencies

Permissions-aware MCP gateway with human-in-the-loop approval for AI agents

Cloud human input handoff for AI agents.

Direct Anthropic provider for loom — runs agents through the Anthropic SDK with prompt caching and idempotent spawns.

Official SDK for the Pump Up agent API — human-in-the-loop approvals, elicitations, and routing for AI agents.

TypeScript SDK for the HAX (Human Approval eXchange) API

Browser MCP — control your real Chrome from Claude Code. 34 tools, CAPTCHA solving, date pickers, autocomplete combobox, overlay dismissal, file upload, multi-session, human-in-the-loop.

TypeScript SDK for creating TryAgent escalations.

EMILIA Protocol MCP Server — trust enforcement for high-risk actions via MCP. Pre-action binding, policy-bound verification, and accountable human signoff.

Local-model provider for loom — runs agents through a local Ollama server, no API key.

gotoHuman - The easiest way to review agent actions and AI‑generated content.

Human-in-the-loop workflows for Loopstack — ask questions, present options, and request confirmations from users during workflow execution

An MCP server that adds a human-in-the-loop step to any AI pipeline by reaching a human on their own messaging app (Telegram first).

Interaction modes for Reactive Agents — 5 autonomy modes, checkpoints, and human collaboration

Human-in-the-loop approval gates for AI agents — pause function calls for human review, then resume or abort.

Zero-dependency native PDF generation library. 22 scripts (Arabic, Hebrew, Thai, CJK, Devanagari, Telugu, Bengali, Tamil, Sinhala, Tibetan, Khmer, Myanmar, Ethiopic, Cyrillic, Greek, Georgian, Armenian, Latin), BiDi, PDF/A-1b/2b/3b, AES encryption, digita

OpenTrain MCP server: gives MCP-capable agents tools to register an OpenTrain account, draft and publish jobs, review proposals, hire freelancers, manage contracts and milestones, send messages, poll updates or register webhooks, and manage credits and te

AI agent command firewall with Telegram-based human approval

AI coding agent orchestrator — resolve GitHub issues end-to-end with isolated git worktrees, quality gates, and an MCP server. Works with Claude Code or Aider.

An MCP server that lets AI agents (Claude, Cursor, etc.) reach you on any channel — desktop, Telegram, SMS, email — with two-way ask/reply, real-time inbox push, Do Not Disturb, idle gating, multi-session routing, and a one-page web UI for setup. Zero con

yunomi 🍵 — Human-in-the-loop approval gate for AI coding workflows. The AI serves its report; you review and approve. (formerly reviw)

QA Foundry - AI-Powered QA Automation Platform. Record tests, generate with AI, run Playwright tests, review with HITL, track results.

Fast TypeScript AI agent framework — per-request agents, 30+ model providers, 100+ integrations, 20+ vector DBs, 10+ databases, sessions, memory, knowledge, tracing, evals, HITL, teams, and workflows.

loom's build-time assembly layer — discovers and reconciles extensions, validates a bundle's declarative shape, and assembles the Registry the kernel ticks. Runs once at start-up, off the replay hot path; the kernel needs none of it to tick.

Registry discovery client for loop definition catalogs.

Nominee adapter for Vercel Eve — policy, approvals, receipts, and fresh tokens for Eve agent tools

WaxSeal MCP Server — verify cryptographic identities, sign documents, and create AI approval gates from Claude Desktop, Cursor, and Windsurf.

Humancy integration plugin for Agency - enables agents to request human input

n8n node to request human reviews in AI workflows with gotoHuman

Official ClearedBy SDK — gate agent actions, wait for human review, guard a function.

n8n community node for LoopQuest — gamified human-in-the-loop review.

Nominee adapter for the Vercel AI SDK — policy, approvals, receipts, and fresh tokens for AI SDK tools (guardTools / nomineeTool)

Entry point for building and running governed loops.

Thin TypeScript client for Cosignet — Critical Action Control with payload-bound approvals.

Agent Tick control-plane integration for Pi: remote decisions, approvals, and session status updates

A persistent AI engineering team you run as CEO: specialists who disagree catch what one pass misses, memory compounds across runs, and you keep the final say

Runtime governance layer for generative AI agents. Works with any MCP-compatible client — Claude, GPT, Gemini, Cursor, or custom frameworks. MAI classification, forensic audit trails, human-in-the-loop gates, EU AI Act compliance, NIST/SOC 2/CMMC mapping.

MCP server for Stacksona Gate that lets AI agents request approvals, log audit events, poll decisions, and validate approval tokens.

LoopPause MCP server — pause agent execution and route human approval requests

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

MCP server that proxies all agent actions, classifies risk, enforces natural-language policies, and logs to a local SQLite audit DB. View activity in the hosted dashboard at roxy-agent-web.vercel.app.

Turn the OpenAI Agents SDK human-in-the-loop tool-approval step into a verifiable, offline-checkable EMILIA authorization receipt. OpenAI pauses an agent and asks for approval; EMILIA makes that approval portable, tamper-evident evidence — a named human a

Open-source encrypted AI agent inbox and agent-to-agent messaging CLI with permanent addresses, durable private threads, signed channels, JSON automation, and human approvals.

The Enterprise Statechart Engine for AI Agents. A hardened, schema-driven orchestration layer built on LangGraph.js.

typescript client for humanlayer.dev

Execution engine for loop state transitions and lifecycle.

ClearedBy MCP server — request_clearance, check_policy, get_ledger from any MCP client.

MCP stdio server for Pileless. Wraps the Pileless REST API in MCP tool calls so Claude Code, Cursor, Cline, etc. can submit Piles natively.

Autonomous AI agents you deploy in 30 seconds across any channel. Zero-config CLI, 37 channel adapters, HEXACO personality and cognitive memory powered by AgentOS, five-tier prompt-injection defense.

The runtime checkpoint for AI agents — inspect every tool call, hold the irreversible for approval, keep verifiable evidence of human oversight.

Canonical loop types, contracts, and model primitives.

MCP server for Orbiter Dev — Approve, reject, and monitor AI coding agent actions from your phone. Works with Claude Code, Codex, Cursor, Windsurf, and any MCP-compatible agent.

n8n community node for Qorami — check an email before your workflow sends it (send / ask a human / block).

TLNT - HMS-Powered Multi-Agent Platform with Government Agency Analysis, Deep Research, and Enterprise-Ready Deployment. Self-optimizing multi-domain AI agent with continuous learning and enterprise-grade performance monitoring.

The live attention layer for Paperclip, combining Aperture Core continuity with Paperclip-native operator policy.

MCP server for LoopQuest — let any AI agent request human-in-the-loop review.

Typed async workflows with automatic error inference. Build type-safe workflows with Result types, step caching, resume state, and human-in-the-loop support.

Approval, confidence, and policy guard framework.

Human-in-the-loop MCP server for AI agents — biometric approval with PKI-signed attestation. stdio + HTTP.

Local MCP server for human.menu with non-custodial x402 payments via agent-pay.

Official TypeScript SDK for Agent Action Firewall - AI agent security platform

MongoDB-native durable workflow orchestration engine. Like Temporal but simpler - supports sleep, wait, retry, parallel execution, human-in-the-loop, and crash recovery. Perfect for payment gateways, approval flows, and scheduled tasks.

The open execution boundary for autonomous AI agents — deterministic gates, hard limits, and tamper-evident action receipts.

Example workflow that uses @loopstack/hitl's AskUserWorkflow to prompt the user for free-text input and echoes the answer back

TypeScript SDK for Nod approval requests and webhooks.

Command-line client for the OpenTrain public API: register an agent account, draft and publish jobs, review proposals, hire, manage contracts and milestones, message, poll updates or register webhooks, and manage credits and your team.

sudo for AI agents — remote approval + signed receipts for MCP and Claude Code tool calls

TellyMCP - Telegram control plane for MCP-connected coding agents

Agentic sales pipeline — detects buying-intent signals on Reddit, HN, and RSS feeds via a managed brain, drafts outreach for review, and (v2.0) sends approved Reddit replies and DMs with your own credentials. The math runs on us, the sending runs on you.

Signal primitives for feedback and training loops.

Example workflow that uses @loopstack/hitl's ConfirmUserWorkflow to present a markdown summary and branch on the user's confirm/deny decision

Runtime guardrails for TypeScript AI agents — prevents duplicate tool calls, enforces per-user LLM cost budgets, and gates irreversible actions before they execute. Works with Mastra, Vercel AI SDK, LangGraph.js, and OpenAI Agents SDK.

Actor attribution model for human and AI actions.

PostgreSQL loop state store for durable production workloads.

Event bus contracts for loop execution signals.

The trusted human layer for AI agents. Call a vetted human from your agent to execute a task or verify an answer, with proof.

Core proxy engine for Cordon for MCP — the security gateway for MCP tool calls

TypeScript reference implementation of the guardian-agent spec: a runtime supervisor for tool-using LLM agents (audit log, tool-permission scoping, HITL approval gates, emergency-stop).

WhatsApp MCP Server: Human-in-the-loop for AI agents via WhatsApp for confirmations, notifications, and interactive questions.

n8n community node for ClearedBy — gate an action behind policy + human approval before your workflow proceeds.

Cloud Chromium browsers for AI agents — legitimate web access with live viewer, one-click human takeover, and a clean residential connection. MCP + A2A ready.

Find the safest first AI workflow before investing in AI agents or automation.

In-memory loop state store for local development.

Perplexity Sonar API adapter for Loop Engine multi-LLM steps (grounded search + reasoning).

HTTP routing adapter for approval and event delivery endpoints.

Public AI operations templates and a starter MCP server for human-reviewed AI operations.

Guard OpenAI-compatible tool calls (OpenAI, xAI Grok, Together, …) with EMILIA Protocol — block irreversible agent actions until a named human signs off.

TypeScript framework for production AI agents with explicit workflows, provider-agnostic models, streaming, interrupts, and runtime persistence.

TypeScript types, schemas, builders, and validation for AgentClutch Action Cards.

MCP Server for TouchGrass — hire Orb-verified humans for real-world tasks

Human-in-the-loop and notifications for the pi coding agent, powered by ping-a-human. One command: npx ping-a-human-pi.

Habena — keep your AI agent on a short rein. MCP middleware proxy: policy guardrails, spend caps, and human approval for AI agents.

Cordon for MCP — security gateway for MCP tool calls. Firewall, auditor, and human-in-the-loop approvals over any stdio MCP server.

VantaGate MCP Server & OpenAPI integration - Human-in-the-Loop for AI Agents

Official CLI for pdfnative — render JSON to PDF (22 Unicode scripts, math symbols, COLRv1 colour emoji, bookmarks/outline, true constant-memory streaming), merge/split/extract pages, annotate (markup annotations), sign (RSA + ECDSA, native constant-time c

Human-in-the-loop example using AG-Kit Agents

Human-in-the-loop approval gates and real-time push notifications for AI agents

Google Gemini adapter for governed AI loop actors.

MCP server bridging AI coding agents (Cursor, Claude Code, Codex, Windsurf, Kiro, Antigravity) to the Customaise Chrome extension. 18 tools: UserScripts, AgentScripts, WebMCP calls with HITL consent, visual DOM targeting, tab control.

Governance primitives for AI-assisted compliance systems: fail-closed HITL (human-in-the-loop) gate and SHA-256 hash-chained provenance records. Framework-agnostic, zero runtime deps beyond zod. (The disclosure-card schema lives in the sibling package @pr

The open-source safety stack for AI agents — one command to make your agent safe

Commerce Gateway framework adapter for governed tool routing.

Headless, model-agnostic runtime for LLM workflows — records state, artifacts, rules, human-review gates, traces, cost telemetry, and tamper-evident run bundles.

n8n community node for IdentArk — route LLM calls through a secure gateway so your n8n instance holds zero API keys

IP Bot CLI (ipbot) — accessible-by-default social publishing for AI agents and humans: alt text on every image, human-in-the-loop approval, agent-grade automation

Pre-execution governance for AI agents. Intercepts MCP tool calls before execution with deterministic blocking, human-in-the-loop holds, and behavioral drift detection.

xAI Grok adapter for governed AI loop actors.

MCP server that proposes Signal messages and groups but sends nothing without explicit human approval via a local web GUI.

Devtools and replay UI for debugging governed loops.

PagerDuty routing adapter for human approval delivery.

Loop lifecycle contracts for AgentClutch action proposals, decisions, events, and resume context.

MCP Server for the Beacon Protocol

MCP server for governed AI workflows — invoice OCR, AML triage, credit review, and custom contracts with human approval gates and audit trails

Core AgentClutch sessions, consequence metadata, risk utilities, lessons, and Run Story helpers.

Human-in-the-loop workflows enabling powerful task management with the Tasks & Queue collections

MCP server for AI agents to browse and book humans on rentahuman.ai

Human-in-the-loop approval SDK for AI agents

Human-in-the-loop policy engine and surface adapters

React-compatible components for AgentClutch Action Cards and takeover UX.

Approve or deny Claude Code permission prompts from your phone — so long autonomous runs never stall while you're away from the desk. Self-hosted, token-gated, over Tailscale or LAN.

Datashift SDK for human-in-the-loop AI agent workflows

React control-surface components (WorkbenchShell) for inspecting and editing LLM Workbench runs — artifacts, rules, traces, gates, and bundle import/export.

A human-approval gate for the dangerous things an AI agent does.

MCP server for NeedHuman — Human-as-a-Service API for AI agents. When your agent needs a real human (accept ToS, create accounts, verify identity), it calls NeedHuman.

Playwright adapter for AgentClutch Action Cards, rules, and takeover UX.

n8n community nodes for Ujex — Postbox, Approvals, and Memory for agent workflows.

Extensible human-in-the-loop interaction toolkit with adaptive prompt UI and flexible decision schemas.

MCP server that dispatches physical-world tasks to humans through a unified API

First-party n8n node for Humangent — a human-in-the-loop inbox for n8n workflows.

HeySummon — Human-in-the-loop for AI agents. Install & run with one command.

Deterministic orchestrator for AI coding agents — zero tokens, full SDLC pipeline, human-in-the-loop gates.

MCP server (+ OpenClaw SKILL.md) that gives AI agents access to real-world people who listed themselves to be hired by agents. 31 tools including search by skill/location/equipment, job offers, job board listings, in-job messaging, streaming payments, and

Human-in-the-Loop MCP Server — enables LLM agents to ask questions across all your devices via ntfy.sh

The deterministic judgment engine inside Aperture.

EP-MCP middleware (reference implementation). Wraps an MCP tool-call handler: irreversible tool calls route through consent → Class-A signoff → an emitted EP-RECEIPT-v1 + an appended provenance entry; everything else passes through. Includes the demand ho

Interactive human-in-the-loop design variant picker for OpenCode. The agent shows HTML design variants in a browser grid; the user picks or asks for regenerations (produced by the same conversation's model) until satisfied.

Native LoopPause human-in-the-loop nodes for LangGraph (TypeScript)

Sanctum Runtime SDK — the runtime trust boundary for autonomous systems. Signed action tokens, blast-radius scoring, source-trust classification (indirect-prompt-injection defense), dual-approver + auto-escalate, policy replay, SOC2 + NIST AI RMF evidence

AI orchestration platform with autonomous teams, enhancement planning, migration tools, 25+ slash commands, checkpoints & hooks. Multi-IDE: Claude, Cursor, Windsurf, Cline, Copilot

Human-in-the-loop approval workflow for tool-use-firewall

CLI for AgentClutch demos and local run inspection.

Shared HTTP utilities for agent-safe action and workflow runner adapters.

API gateway for AI agents with human-in-the-loop write approval

A tiny TypeScript approval gate for AI tool calls, MCP tools, and human-in-the-loop workflows.

Carbon-based node orchestration framework — humans as distributed worker nodes, with AI planning, team management, and performance evaluation

Human-in-the-loop example using Mastra

Guard LangChain.js tools with EMILIA Protocol — require an offline-verifiable authorization receipt (EP-RECEIPT-v1) before an irreversible tool runs: missing -> refused, valid -> runs, replay -> refused, forged -> refused (RR-1). The approval becomes port

Agent-safe action and workflow runner core for TypeScript backends with schema validation, approval, and audit.

harmony-agentic-decisions. Production-grade human-in-the-loop decision workflows with LLMs and messaging channels.

Cheqpoint MCP server — human-in-the-loop approval tools for Claude, Cursor, and any MCP client

ProxyHuman MCP server — ask a human for help with a browser-based task (sign-in, captcha, OTP, payment, subjective choices). Includes the proxyhuman CLI for signup + config.

MCP server exposing ConsentGate consent checks and human-in-the-loop approvals as agent tools.

MCP server for patxin.com — patch in a human for real-time cognitive micro-tasks

The first Ledger hardware-wallet plugin for the Hedera Agent Kit — a human-in-the-loop signer that Clear-Signs high-risk EVM transactions on a physical Ledger before a Hedera agent executes them. The device is the final gate.

A Codex skill to plan and run the hybrid-workforce operating model: workflows made of loops, run by swappable agents and humans, gated by verifiers, improved by corrections.

Human intelligence infrastructure API for AI agents. Verify facts, review emotional responses, and request real-world observations from human domain experts.

Claude Agent SDK human-in-the-loop example

Inline review tool for MDXG markdown — capture highlighted comments and export them as structured JSON for LLM agents.

Blax - HMS-Powered Multi-Agent Platform with Government Agency Analysis, Deep Research, and Enterprise-Ready Deployment. No local LLM keys required.

Pre-execution governance for AI agents. Intercepts MCP tool calls before execution with deterministic blocking, human-in-the-loop holds, and behavioral drift detection.

A lightweight, database-native AI agent workflow engine. 让 AI 长任务不再跑丢，每一步都能看见，失败可以从断点继续。

Cerberus — a local-first security gateway for autonomous AI coding agents. Intercept, risk-score, and human-approve every tool call. Terminal-first, with a forensic dashboard.

Declarative governance and policy enforcement for Vercel Eve agents — deny, escalate, audit, and allow tool calls with composable rules, OWASP Agentic Top 10 coverage, and tamper-evident audit trails.

Qorami client — a control point between your AI agents and sending email (send / request_human_confirmation / do_not_send).

protectAgent() for Sanctum Runtime — verify-before-execute wrapper for AI agents, tool use, and LLM automations.

Generflow renderer — React, Web Components, and vanilla TS. Progressive UI from low-token GF-Lang specs streamed over SSE.

MCP exporter for Agent Action Runner.

Pure approval state machine and local pending store/channel for human-in-the-loop agent gates.

CLI for Runhuman - AI-orchestrated human QA testing

The Civic Access Protocol kit: the code embodiment of The Open State Constitution. Encrypted on-device session vault (Art. 1), the two-phase human-confirm gate (Art. 2), and citizen-driven browser session capture (Art. 10) — shared by every Open State imp

n8n community node for OKRunit human-in-the-loop approval gateway

MCP server for delegating tasks from AI agents (Claude, GPT, etc.) to real human virtual assistants. Human-in-the-loop AI agent handoff via the TaskBullet API.

Human-in-the-loop example using AG-Kit Agents

Human-gated test attestation system with cryptographic signing

NestJS adapter for agent-safe action and workflow runner for TypeScript backends.

CLI for Agent Action Runner local development, manifest inspection, workflow validation, and MCP previews.

LangChain TypeScript SDK - Human-in-the-Loop validation via Loopman platform

IP Bot MCP Server — accessible-by-default social publishing tools for AI agents (Model Context Protocol): alt text as a first-class param, human-in-the-loop approval

MCP server for HumanSurvey — let AI agents collect structured feedback from groups of people

JamJet engine — policy evaluation, budget enforcement, audit writing, approval queue. The shared brain across all JamJet adapters.

Express adapter for agent-safe action and workflow runner for TypeScript backends.

MCP server for AI-native forms, feedback forms, surveys, and human answer collection.

TypeScript client and CLI for the EarnFi Agent API — registration, x402 USDC paid job creates, polling, and creator actions.

MCP server for Verified Workflows — human-in-the-loop review as AI agent tools

📎 A Windows-98 Clippy that narrates your Claude Code session, gates dangerous commands, and answers your questions — without touching your terminal.

AI-native web browser runtime for autonomous agents with human supervision

Production-ready TypeScript patterns to build solid and robust AI applications. Retry logic, circuit breakers, rate limiting, human-in-the-loop escalation, prompt versioning, response validation, context window management, and more—all with complete type

A versatile JavaScript library for designing and managing intelligent agents capable of interacting with AI-driven chat models, leveraging OpenAI's GPT for dynamic conversations and task automation.

n8n community node for Reeflex - deterministic governance for AI-agent actions. Calls reeflex-core POST /v1/decide and routes each item to Allowed, Held for Approval, or Denied.

Human-approval UX for AI agents and CLIs: explain risky actions in plain language before a human approves them.

HITL infrastructure for AI agents. Your agent calls awaitHuman(), a human reviews via Slack/email/dashboard, agent resumes with a typed response.

MCP Server for ORACLE-H — Human-on-the-Loop validation for autonomous AI agents

A React component library for building AI chat interfaces with LangChain/LangGraph integration.

An MCP server that turns AI-supplied JSON into an interactive local web form, returns structured answers, and keeps a browsable history dashboard.

Context-Driven Development - AI-native productivity workflow with human-centered decision making

CHIP + Atom of Thoughts (AoT) Agent Harness — multi-provider n8n community node with a Process Analyst Worker for operational case triage (email, document, automotive service). Pluggable vaults (Obsidian, Supabase) and Human-in-the-Loop gate.

Human control plane for agent execution — queue, plan approval, audit

Embeddable execution surface for ageflow workflows: stream, fire-and-forget, async HITL, cancellation.

Dynamic Integration between AI and Labor - A coordination framework for AI and human specialists making decisions together within state machines

TypeScript SDK for the LoopQuest human-in-the-loop review API.