{"slug": "designing-ai-with-character-what-we-learned-building-berd", "title": "Designing AI with character: what we learned building Berd", "summary": "Block Inc. has open-sourced Berd, a desktop application for working with AI agents across projects, skills, tools, and models, designed to give agents distinctive visual identities to make them more approachable. The tool grew from Block's experience with its goose agent framework, which it introduced in January 2025, and Block joined Anthropic, OpenAI, and others in December 2025 to establish the Agentic AI Foundation under the Linux Foundation.", "body_md": "At Block, we often build the tools we need ourselves. [Berd](https://berd.xyz/) is a desktop application our teams use to work with AI agents across projects, skills, tools, and models. Today, we’re making Berd open source, sharing the application, and the design and technical lessons behind it for others to inspect, use, and adapt.\n\nBuilding Berd also gave us a chance to explore a design question: How do you make something as abstract as an AI agent easier to understand and shape? Our answer was to give agents character, not only through roles, instructions, skills, and tools, but through distinctive visual identities.\n\nOur approach builds on Block’s long history of bringing inspired design to categories not traditionally known for it. Square transformed payment hardware from something sellers hid behind the counter into something they could display proudly. Cash App brought personality and cultural relevance to personal finance. In each case, design changed how people related to it, and who felt invited to use it. With Berd, we brought the same instinct to AI. The goal was not to disguise the technology, but to make its capabilities more visible, approachable, and personal.\n\n**Built for the way we work**\n\nBerd grew from a practical problem inside Block. We had access to increasingly capable agents through tools such as goose, Claude Code, and Codex, but working with them meant navigating different interfaces, configuration systems, and ways of managing context. The technology was powerful, but the experience around it was fragmented, and often assumed a level of technical fluency that many people did not have. We learned that we didn’t need another model or agent harness, we needed a consistent environment around them.\n\nBerd gave our teams one desktop application for working across models and harnesses. It brought conversations, files, folders, instructions, agents, and skills together around persistent projects. Instead of rebuilding context for every task, people could shape agents around recurring ways of working and return to that work later.\n\nThis was designed to make agentic work more approachable beyond engineering. People could begin with a conversation, understand which context and tools were active, and add more structure as the work required it.\n\n**Giving AI more character**\n\nMany AI interfaces begin with an empty prompt box. The model may be capable, but the product gives people little sense of how the agent is configured, which context and tools are available to it, and how it differs from another agent. We wanted to make those distinctions more visible and approachable for anyone to build.\n\nIn Berd, character is both visual and functional. Different agents can look distinct because they are distinct. They can help you expand your thinking or narrow down on a solution, write in your unique style, or build your own tools using Berd. Just by chatting with Berd, users can easily create custom agents to do any task they want, even helping with everyday tasks like planning travel or shopping.\n\nTo help people build and share their agents, we’ve designed distinctive collections of animated characters, including our flagship “Gloopies.” These designs give abstract configurations a recognizable visual identity that people can understand and have fun with.\n\nMore than just giving a playful twist to the agent workflow, this approach makes creating and customizing personal agents more accessible for anyone. The avatars make the agent recognizable. Its role, skills, and tools make it useful.\n\n**From goose to Berd**\n\nBerd grew from what we learned building and using [goose](https://github.com/aaif-goose/goose), the open source AI agent framework Block introduced in January 2025. goose connects language models to tools and real-world actions through an open, modular architecture. In December 2025, Block joined Anthropic, OpenAI, and others to establish the [Agentic AI Foundation](https://aaif.io/) under the Linux Foundation. We later contributed goose to the foundation, giving it a vendor-neutral home alongside the Model Context Protocol and [AGENTS.md](http://agents.md).\n\nBerd and goose now serve different but connected purposes. goose remains the open agent framework and runtime. Berd is a desktop application built around it.\n\nBerd connects to goose through the Agent Client Protocol. This separation lets Berd focus on the desktop experience, including projects, context, sessions, agents, and configuration, while goose handles the underlying agent loop.\n\n**Solo work to shared work**\n\nBuilding Berd also helped us understand where a private desktop experience stops. Work with an agent often begins alone. You research, experiment, gather context, and shape an idea before it is ready for other people. Berd gave us a place to explore that individual experience.\n\nBut work rarely stays private. Eventually, you may need to bring in a teammate, add another agent, share an artifact, explain a decision, or keep a record of how something was made. Those are collaborative problems, and they are the focus of our recently released [Buzz](https://buzz.xyz/).\n\nBuzz is our open source workspace where people and agents work together in shared rooms, on a relay the community can control. People and agents have their own identities, participate in the same conversations, and contribute to the same searchable record. What we learned from Berd will inform our continued work on Buzz. Berd showed us the importance of private space, durable context, recognizable agent identities, reusable skills, visible configuration, and clearer visibility into an agent’s configured context, tools, and capabilities. Buzz gives those ideas somewhere to go when individual work becomes collaborative. Start alone, then go multiplayer.\n\nAs always, we want to hear from you. Download [Berd](https://github.com/block/berd), try it with your own agents and workflows, and share your feedback. We want to see what you make, and hear what works for you, and what you think agent experiences should become.", "url": "https://wpnews.pro/news/designing-ai-with-character-what-we-learned-building-berd", "canonical_source": "https://block.xyz/inside/designing-ai-with-character-what-we-learned-building-berd", "published_at": "2026-08-19 05:26:26+00:00", "updated_at": "2026-08-19 05:40:54.160709+00:00", "lang": "en", "topics": ["ai-agents", "ai-products", "ai-tools", "generative-ai", "ai-research"], "entities": ["Block Inc.", "Berd", "goose", "Anthropic", "OpenAI", "Agentic AI Foundation", "Linux Foundation", "Square"], "alternates": {"html": "https://wpnews.pro/news/designing-ai-with-character-what-we-learned-building-berd", "markdown": "https://wpnews.pro/news/designing-ai-with-character-what-we-learned-building-berd.md", "text": "https://wpnews.pro/news/designing-ai-with-character-what-we-learned-building-berd.txt", "jsonld": "https://wpnews.pro/news/designing-ai-with-character-what-we-learned-building-berd.jsonld"}}