Block Open Sources Berd, a Desktop Environment for AI Agents
A single desktop workspace for AI agents working with local files.
1Block's New Open Source Project
Block, the technology company founded by former Twitter CEO Jack Dorsey and owner of Square, Cash App, and Tidal, has open-sourced Berd, a desktop application designed to provide a unified environment for working with AI agents across different models, tools, and projects. Berd is available on GitHub under the Apache 2.0 license, with free downloads for macOS, Windows, and Linux.
2Berd's Core Features
Berd is a locally installed graphical desktop application, not a browser-based workspace. Users can start chats, attach files or folders, select AI agents and models, work within projects, and create automations. The application ensures transparency by allowing users to see which project, file, agent, and model are actively shaping the conversation.
3Block's AI Strategy
Block developed Berd to simplify AI workflows. The company aims to make AI agents accessible beyond engineering fields. Berd allows users to start with a conversation and add tools, context, and structure as needed. This approach seeks to broaden the adoption of AI technologies.
4Berd's Technical Infrastructure
Berd is built with Tauri 2 and React 19, web technologies for lightweight desktop apps, and integrates with Goose, Block's open-source AI framework introduced in January 2025. Goose manages the core agent loops, while Berd handles projects, sessions, context, and configurations. This architecture enhances AI workflow efficiency.
5Berd's Visual Identity
Berd assigns recognizable visual identities to AI agents, helping users distinguish between them easily. Block created collections of animated characters, showcased on Berd's site, to represent agent roles and capabilities visually. This approach enhances user interaction with AI agents.
6Berd's Enterprise Benefits
Berd makes AI workflows more accessible for enterprise users. Persistent projects allow users to return to established collections of files, instructions, and agent configurations, eliminating the need to reconstruct context for each task. This boosts productivity and efficiency.
7Berd's Mobile Strategy
Block does not position Berd as a mobile product. Instead, the company prefers its open-source collaboration platform, Buzz, for mobile experiences. Buzz is optimized for on-the-go work, reflecting Block's belief that AI agents should look different in mobile and desktop environments.
8Berd's Future
Block aims to make AI workflows more accessible and efficient with Berd. Being open-source, Berd can reach a wide user base and contribute to the widespread adoption of AI technologies. Block plans to shape Berd's future development based on community feedback.
9Berd's Community Contribution
Berd is actively developed by a community of 91 contributors on GitHub. The application has reached version 0.6.2, marking its seventh public release. Community contributions accelerate Berd's development and ensure it meets user needs more effectively.
10Berd's Role in the AI Ecosystem
Berd positions itself as an orchestration layer in the AI ecosystem. It is not a new foundational model or agent runtime but aims to make existing AI tools more efficient. This approach facilitates broader adoption of AI technologies.
11Berd's User Experience
Berd enables users to work more effectively with AI agents. The application visually represents agent roles, capabilities, and tools, helping users manage workflows better. This transparency enhances the efficiency of AI workflows.
12Berd's Impact on AI Workflows
Berd has the potential to significantly impact AI workflows by making them more accessible and efficient. Block's open-source strategy could drive broader adoption of AI technologies. Berd may play a crucial role in shaping the future of AI development.