Berd Review 2026

Berd is Block's open-source desktop agent workspace for organizing projects, chats, files, models, tools, skills, and subagent activity in a model-neutral environment powered by Goose.

What Is Berd?

Berd is a desktop command center for agent work. Built by Block and backed by Goose, it gathers project context, conversations, files, skills, tools, models, and subagents into one interface.

Berd is more model- and harness-neutral than an AI IDE and more general than a session manager dedicated to one coding agent.

Workspace

• Project organization
• Persistent agent chats
• File and context management
• Unified desktop interface

Models and Agents

• Goose-powered agent backend
• Multiple model providers
• Subagent activity visibility
• Custom agent workflows

Extensibility

• Skills and tools
• Local project access
• Open-source Tauri and React application
• Developer-controlled configuration

How It Fits an Agentic Development Workflow

Berd is most useful when its permissions, repository scope, model access, and review checkpoints are configured before the first large task. Start with a bounded project, verify the generated changes or analysis, and only then expand access. This matters because agentic tools can move faster than a conventional assistant and can also amplify a bad assumption.

For teams, the practical evaluation should cover setup time, context quality, task completion rate, review burden, model or infrastructure cost, security controls, and how easily a human can interrupt or redirect work.

Best Use Cases

• Developers wanting an open desktop agent command center.
• Teams experimenting with model-neutral agent workflows.
• Goose users who prefer a project-oriented visual workspace.

Privacy, Security, and Deployment

Berd is local and open source, but model prompts and tool calls follow the configured Goose providers and extensions. Review API retention, local data, credentials, and system permissions.

Any coding agent or developer tool with repository, terminal, browser, or secret access should be introduced with least-privilege credentials, protected branches, mandatory review, isolated test environments, and clear log-retention rules.

Limitations

• Early open-source software may change quickly.
• Users still configure models, keys, and permissions.
• The workspace does not replace a full IDE or repository platform.
• Multi-agent workflows need disciplined resource and context management.

Alternatives

• Cindy Agent
• Superset
• Conductor
• Warp
• Orca

The best alternative depends on whether the priority is an AI-native editor, a terminal harness, autonomous cloud execution, code review, mobile control, or a shared orchestration layer.

Verdict

A promising open-source agent workspace that captures the shift from single coding sessions toward a desktop control plane for many models and agents.

Official Sources

• Official Block GitHub repository