# Project Orchestrator

A knowledge-graph backed platform that lets multiple AI coding agents share code understanding, plans and decisions through a central Neo4j-based knowledge base, with semantic search and MCP tool integration for Claude Code and other clients.

Project Orchestrator gives AI coding agents a shared knowledge base instead of having each one start from scratch. Code structure, plans, decisions and progress are stored in a Neo4j graph database and made searchable by meaning through Meilisearch, so multiple agents working on a project can draw on the same understanding.

It provides a plan and task management system with dependencies and progress tracking, a protocol finite-state-machine engine for repeatable workflows, an RFC lifecycle for architectural decisions, and Tree-sitter based parsing for Rust, TypeScript, Python, Go and other languages. It exposes MCP tools for Claude Code, OpenAI Agents and Cursor, supports Google OAuth2 and OIDC authentication, and can run autonomous multi-agent plans with parallel task dispatch.

It ships as a desktop application for macOS, Windows and Linux, with real-time updates over WebSocket and NATS for multi-instance deployments.

Website: https://github.com/this-rs/project-orchestrator