Havuz

The gate between an AI agent and real money. Runtime authorization for financial AI agents: structured intent, deterministic policy, human a

TypeScript AI agent runtime. Write your agents in code and choose how predictable they are: deterministic steps where they can be, an LLM on

Rules-and-receipts layer for AI agents that spend money: plain-English policy → human approval → signed receipts anyone can verify.

👁️ The visual layer for CLI coding agents. Pipe JSON → interactive native UI → modified JSON back. Like fzf, but for visual decisions.

Stateless governance proxy for MCP (2026-07-28): policy, dry-run, human confirmation, audit, tracing

Generative UI for working with coding agents. Review, annotate, brainstorm, and give feedback through task-specific desktop workbenches.

Build and supervise scheduled AI workers with an open-source runtime, human approvals, run logs, REST API, and MCP tools.

NOW you can let AI agents make code changes without losing engineering control.

TypeScript SDK for durable agents, workflows, and custom execution kinds. Built by Swirls.

Let your AI agents call and text your actual phone (iPhone or Android) — Claude Code plugin + agent skill

Local-first Agent Skills for safe social-web workflows—starting with Douyin and designed around reusable browser adapters.

Extracts claims from SOPs, policies and regulations, has a person approve or reject each one, then serves the approved set to agents over MC

Desktop personal AI agent for local projects, coding and learning. Persistent conversations, editable memory, scoped helpers and a living Vi

The live attention layer for Paperclip. Re-stacks approvals, issues, and agent signals into now, next, and ambient.

A Sovereign Agent Governance Protocol for Deep-Water Workflows

Personal, endpoint-encrypted CAPTCHA handoff between an Agent and Android

Policy-aware AI permission reviewer for OpenCode V1 and V2. Evaluates tool and command requests, allowing, denying with feedback, or escalat

Evidence-grounded AI secretary for Claude Code and Codex, with frozen context, deterministic retrieval, bounded quality review, and explicit

Build and operate production-safe agents that run in your own environment.

Read-only trading journal and review harness: Jev typed judgments, agent integration, and a reproducible finance benchmark. No orders, no ad

Mail for AI agents ( and humans). MCP server giving Claude (or any agent) safe, permission-controlled access to email (Microsoft Graph + IMA

Inline review comments for markdown specs. Built-in MCP server hands feedback directly to your AI agent.

Self-hosted, cross-platform control plane for persistent AI employees with local channels, replaceable Worker Hosts, human approval, memory,

Free Claude Code plugin that surfaces buyer-intent Reddit threads with no API key and no Reddit account. Reads public RSS feeds, scores post

Evidence-gated Obsidian brain for bounded multi-agent improvement loops.

AI code-review cockpit: Claude reviews PRs into local artifacts; nothing reaches GitHub until you say so

The whole eve stack. On your own machine. Durable Postgres sessions, a Docker sandbox, memory, approvals, schedules, 1,070 tool integrations

Druks runs agent apps on your own infrastructure: durable runs that survive restarts, human approvals, and full history. Apps install as Pyt

Local-first Agent communication runtime and gateway for A2A, MCP, E2EE messaging, REST/Webhook, and multi-agent collaboration.

Local-first policy gateway for coding agents. Allow, review, or block MCP, shell, SQL, GitHub, and HTTP actions with scoped credentials and

Talk to your computer. Icebreaker transforms natural language into safe Linux actions through isolated AI brains, kernel-enforced sandboxes.

Open-source, self-hosted AI employees that ask before they act. 46 business roles, approval-gated autonomy, untrusted-content safeguards, sa

影武者 / kagemusha: a Markdown governance kit for Claude Code and AI agents—approval queues, correction promotion, and judgment distillation.

An MCP server that lets AI agents operate a multi-cluster Kubernetes fleet through an Open Cluster Management hub, with policy, approval, an

Local Codex control with bounded runs, crash recovery, and auditable results. Optional ChatGPT bridge.

Research Pilot helps AI agents understand, maintain, and advance research projects through papers, claims, evidence, experiments, and human-

Open-source governance infrastructure for AI-assisted legal work

AI-automated SDLC pipeline with human-controlled quality gates, for Claude Code — spec before code, evidence before pass, a human sign-off a

Privacy-first document intelligence engine — parse PDFs, DOCX, PPTX, XLSX & CSV into AI-ready chunks for RAG pipelines. Includes HITL review

Local-first AI project OS that turns coding models into reliable software agents with memory, sandboxed execution, verification, recovery, G

Human-in-the-loop multi-agent task orchestration platform for Git-based software workflows.

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 o

Pause your AI agent. Ask a human. Resume with their answer. Open source human-in-the-loop (HITL) library for production LLM agents: Slack, e

AI coding agent that shows its work — every step visible, every patch approved by you. Build projects brick by brick, stay in control.

FinishUP — local-first structured work evidence, deterministic checks, and honest verification receipts.

Arcgram — an interactive diagram engine for human–AI collaboration: an AI agent draws its reasoning as an auditable node-and-edge flow and s

Local-first permission gateway for AI agents with scoped connector access, human approval, audit trails, and encrypted local credentials.

A production-ready toolkit to accelerate and automate the end-to-end lifecycle of AI Agent development.

The open interoperability layer for AI agents. Connect your favorite agents and harnesses. Work together.

Codex Pro Max helps you keep working with Codex without interrupting the same conversation.