What is Clawboo?
Clawboo is an open-source studio for teams of AI agents working on development, marketing, research and personal tasks. Its team lead, Boo Zero, divides requests into jobs, assigns them to teammates and combines their results in chat. Each teammate can use a different supported runtime while sharing a persistent task board and team memory.
The application runs on the user's computer and opens in a browser. Its built-in runtime can use connected AI providers or local models through Ollama; external agent runtimes are optional. Task state is stored in SQLite, and API keys are kept in an encrypted vault. Clawboo requires no account and sends no telemetry itself, but prompts reach the providers and applications connected by the user. The software is free, with model usage billed separately by the selected provider.
The interface includes a live team graph, a calendar-based task board, scheduled routines and runtime health views. MCP connectors provide access to connected applications on a per-agent basis. Budgets warn by default, with hard caps available as an option. Approval controls cover calls routed through Clawboo, not every built-in action performed by external runtimes.
Clawboo screenshots
Explore the interface. Open a screenshot to view it at full size.
Source: official Clawboo project page.
Key capabilities
- Mixed-runtime teams Coordinates built-in agents, Claude Code, Codex, Hermes and OpenClaw on a shared board and in team chat.
- Persistent task tracking Records delegations as task cards, with calendar views, approvals and work requiring human attention.
- Shared memory Lets teammates save and retrieve facts across runtimes through a browsable, searchable memory graph.
- Scheduled routines Sends recurring work to a team chat or assigns a task directly to an agent.
- Application connectors Shares MCP tools and connected applications with selected agents, with approval controls for managed calls.
- Code task verification Uses per-task git worktrees and repository verification commands for qualifying code tasks, with additional review for delegated or large changes.
Where Clawboo fits
Clawboo fits workflows in which specialist agents divide a larger request, such as preparing launch materials, comparing competitors or completing repository tasks. Ready-made teams provide a starting point, while the shared board and chat expose how work passes between agents. Routines can also support recurring briefs and other scheduled assignments.
Before deployment, operators should check runtime prerequisites, provider credentials, application permissions and spending limits. Claude Code requires a global installation with its agent SDK rather than a one-off npx run. External runtimes can execute their built-in tools without Clawboo approval prompts, and a git worktree is not a security sandbox. Dashboard access requires a token before binding beyond localhost, and repository verification applies only to the documented code-task paths.
Install
npx clawboo@latestReviews of Clawboo
Be the first to share your experience with Clawboo.
Write a review
Loading review form…
Alternatives to Clawboo
Hermes AgentMIT-licensed agent from Nous Research that runs in a terminal, desktop app, browser or chat apps. It keeps curated memory, writes and refines its own skills, schedules jobs and works with OpenRouter, Anthropic, OpenAI, local or Nous Portal models.