#human-in-the-loop

VERDICT — a DFIR agent (Claude Code as the engine) that produces a signed, offline-verifiable verdict. SANS Find Evil! 2026.

Open-source loop engineering for auditable business briefings: claims, evidence, gates, findings, repairs, and human review.

Assign real work, share context, review output, and keep every human-agent handoff visible.

Local-first mission control for your AI agent workforce — run CLI agents like employees (SquadCue)

Proxilion is the security layer for the agentic workforce. It turns managed AI agents into governed users by enforcing strict cryptographic boundaries on every API call to SaaS like Google Workspace, Salesforce, or Atlassian.

Playbook for PMs shipping AI products with PRDs, evals, HITL, launch gates, cost, and observability.

Policy-mapped evaluation lab for AI assistant behavior: approval gates, refusal boundaries, uncertainty handling, tool-use grounding, traces, and quality gates.

Agent + Multimodal RAG 全栈学习项目：LangGraph PAE/ReAct、Multi-Query、BM25 + pgvector 混合检索、RAG-Fusion/RRF、MinerU/OCR/ASR、evidence 引用与 HITL；React + Spring Boot + FastAPI。

Reusable Interpretable Context Methodology (ICM) workspace template for filesystem-based AI agent workflows

Multi-Agent AI research platform using LangGraph for automated AI research. Orchestrate specialized agents (Planner, Researcher, Writer, Human-in-the-loop). Features a FastAPI backend, parallel web/academic retrieval (Tavily/Arxiv), and a React frontend for generating cited, grounded AI answer.

Architectural taste for TypeScript coding agents. Minimize conceptual surface area, create boundaries deliberately, and generate code humans can navigate, review, and evolve.

Skein is an agentic AI platform that answers questions by weaving together structured and unstructured data, grounding every response in a knowledge graph, hybrid vector search, and a governed semantic layer. It escalates to humans when unsure, and runs free locally or scales to the cloud.

Agent-safe action and workflow runner for TypeScript backends.

Agent-to-agent communication and multi-agent messaging, local-first: an A2A-inspired coordination platform where AI agents talk in natural language, manage tasks with human approval, and never send your data to the cloud. 65-command API, Python client, web UI, encrypted backups.

Terminal-first VS Code control plane for human-governed AI-assisted software delivery.

Human-in-the-loop planning plugin for Claude and Codex: turn ideas into reviewed Markdown plans, record decisions, enforce explicit execution gates, and provide tested controller primitives for safer agent workflows.

Open-source enterprise AI agent platform for centrally governing agents, tools, memory, RAG, approvals, and durable workflows with FastAPI, LangGraph, LangChain, and LangSmith.

Spec-driven development workflow for Claude Code — built around an approved design doc, where each stage waits for your approval.

AI-powered Statement of Work drafting tool with a Human-in-the-Loop review gate, built with React and the Anthropic Claude API.

OpenAI/Anthropic API-compatible LLM powered by YOU — 人間を事後学習する訓練環境 (an AI trainer post-trains the human)

Reusable agent skills for evidence-backed research, approval-gated publishing, schema-validated records, and durable context handoffs.

Your agent ships. You stay dangerous. A demanding code coach for Claude Code and Codex.

A privacy-aware multi-agent career management system with structured source governance, human approval workflows, and LLM output evaluation.

Technical case study of a production AI-powered lead qualification system built with WhatsApp, Claude, Whisper, Supabase, and Cloudflare Workers.

A Rust-first CLI agent built for real work: durable sessions, multi-provider runs, governed tools, modular workflows, and a human checkpoint before software can change your system.

AI-assisted collections & back-office automation console: risk scoring, human-approved workflows, and audit trail for overdue accounts. Next.js 15 + TypeScript.

Governed AI Agents for State & Local Government on AWS — 8 agentic solutions, MCP authorization gateway, Whole-of-Government orchestration, and compliance-first governance framework

Turn Microsoft Copilot (Excel + Chat) into a disciplined, evidence-driven project operator. 43 skills, frozen acceptance criteria, deterministic lint gate, revision-controlled workbooks. No Cowork, no add-ons — built for strict corporate environments.

Train and run evidence-first AI coding agents with human approval gates, isolated patches, and independent verification.

File-native judgment practice for partner-coupled agents - propose then accept, thin load, your vault stays private

Semi-autonomous issue-to-PR pipeline with cost-ordered routing across local and cloud LLMs.

Multi-agent AI backend built on LangGraph — Supervisor routing, Human-in-the-Loop approvals, Gmail & Calendar tools, real-time SSE streaming, and persistent Postgres-backed memory.

Your agent asks one question at a time until it actually understands the job. A Claude Code skill for people who drive agents without writing code.

Talk with your agent over documents, presentations, diagrams. Leave comments on them and your agent can read them back. A local-first MCP server + live viewer. 🚧 WIP

Agentic back-office ops copilot for financial services — chat an intent, the agent plans and acts on the reconciliation-breaks queue, a human approves, and every outcome is proven against the ledger. Guardrails enforced in code.

Regulatory-change remediation copilot for financial services — the agent maps, groups and drafts; a human approves every cohort; every action is proven and audited. DORA demo.

A secure, extensible agent-runtime CLI for terminal workflows, with deterministic permissions, human approvals, file tools, shell execution, and durable sessions.

A task board AI agents volunteer for — and a human gate they can't get past. Pull-model coordination substrate for agent fleets (Go + SQLite).

Switchback Governance · 折返治理 — the human-in-the-loop governance layer for multi-agent teams (GOAI 2026 Track 1)

NOA — the Agent Action Receipt. Open governance layer for AI agents: gate risky actions before they run, emit a tamper-evident receipt anyone can verify offline. Early access · Apache-2.0.