Show HN: Clawe – open-source Trello for agent teams
Clawe is an open-source 'Trello for agent teams' that provides a coordination layer for multi-agent workflows, moving beyond simple scripts to offer visibility and control. It allows developers to deploy and manage a squad of AI agents with distinct roles, scheduled tasks, and a Kanban-style dashboard. This project addresses the growing need for better orchestration and oversight in the burgeoning field of autonomous AI agents.
The Lowdown
Clawe is an open-source, multi-agent coordination system built on top of OpenClaw, designed to bring structure and visibility to AI agent workflows. The project aims to solve the problem of managing autonomous agents, which often turn into opaque cron jobs and log-tailing exercises, by treating them as long-lived, stateful workers with explicit responsibilities.
Key features and concepts include:
- Multi-Agent Orchestration: Deploy a team of AI agents, each with unique roles, identities, and workspaces.
- Kanban-style Dashboard: A web interface provides a visual overview of squad status, tasks, and allows for direct chat with agents.
- Scheduled Workflows: Agents wake on cron schedules to check for new work, with heartbeats staggered to avoid API rate limits.
- Task Management: Supports assignments, subtasks, and real-time notifications for task updates and @mentions.
- Collaborative Workspaces: Agents share context through a shared file system and a Convex-backed backend.
- Extensible Architecture: Allows for customization, including adding new agents and modifying heartbeat schedules.
- CLI for Agents: Agents interact with the system via a dedicated CLI for checking notifications, listing/updating tasks, and registering deliverables.
- Pre-configured Squad: Ships with four agents: Clawe (Squad Lead), Inky (Content Editor), Pixel (Designer), and Scout (SEO), demonstrating its utility for content review and maintenance workflows.
Clawe seeks to provide a more intuitive and manageable way to operate AI agent systems, transforming them from black boxes into observable, collaborative entities, particularly useful for agencies automating content-related tasks.