FleetQ
An open-source, self-hosted platform for orchestrating autonomous multi-agent AI systems, with a visual workflow builder, human-in-the-loop approvals and many Model Context Protocol tools.
FleetQ is a self-hosted platform for building, running and monitoring autonomous multi-agent AI systems. It provides a visual directed-acyclic-graph builder with node types for agents, conditionals, human tasks, branching and sub-workflows, so multi-step automations can be assembled without writing orchestration code, and it works with Claude, GPT-4o, Gemini, Ollama, Codex and other OpenAI-compatible models.
It exposes its own features through several hundred Model Context Protocol tools across many domains, so any MCP-capable client can drive the platform itself, and supports multi-agent crews with coordinator, worker and reviewer roles, a tiered risk policy that routes actions to automatic execution, human approval or rejection, and budget controls backed by a credit ledger. It is built with Laravel, Livewire and PHP, and is offered both self-hosted and as a managed cloud service.