Clawe is a multi-agent coordination system that enables teams of AI agents to work together with distinct roles, shared workspaces, and scheduled execution.
The tool addresses the challenge of orchestrating multiple AI agents toward collaborative goals. It provides a Kanban-style task board where agents can be assigned work, coordinate through shared files and a backend, and receive notifications of updates and mentions. Each agent operates with its own identity and workspace, waking on cron schedules to check for assigned tasks. The system includes a web dashboard for monitoring agent status, viewing tasks, and communicating with agents directly.
Clawe suits projects that need multiple specialized AI agents working in concert rather than a single agent handling all work. It comes preconfigured with four agents covering different roles—squad lead, content editing, design, and SEO—though new agents can be added by creating workspace templates and updating configuration files. The tool requires Docker, a Convex backend account, and an Anthropic API key to run. It is built on OpenClaw and uses TypeScript throughout.
The project maintains active development with regular updates to both the core coordination system and the web dashboard. The codebase is organized into distinct applications for the squadhub gateway, notification watcher, and dashboard frontend, with clear separation between agent workspaces and shared infrastructure. Documentation covers customization paths including adding new agents and adjusting heartbeat schedules, indicating the maintainers expect users to extend the system for their specific needs.
