Block’s new open-source AI agent does everything from writing code to ordering dinner

Block’s new open-source AI agent ‘codename goose’ has debuted with the ability to write code, handle daily tasks, and adapt capabilities mid-session through a flexible connection framework.

Key features and capabilities; The new AI agent, released under Apache License 2.0, offers seamless interoperability between user interfaces, language models, and various systems through Anthropic’s Model Context Protocol (MCP).

• Users can specify their preferred large language model (LLM) and add new tools during active sessions
• The agent can autonomously execute tasks including writing code, running tests, and managing dependencies
• Goose demonstrates impressive efficiency, with the ability to generate 70% of its own native application code
• Developers report time savings of up to 20% on regular tasks, with some projects completed autonomously in 1-2 days

Technical architecture and improvements; Block completely rewrote the agent, transitioning from Python to Rust to enhance embeddability and performance.

• Integration with Anthropic’s MCP enables connection to common applications like email, Google Calendar, and Drive
• The framework supports both individual and commercial use cases
• Users can connect to data stores, developer spaces, and business applications seamlessly
• The open-source nature allows for community contributions and extensions

Practical applications; Block envisions goose expanding beyond developer tools to serve various business and personal needs.

• Designers can prototype front-ends without engineering support
• Within Block’s ecosystem, goose could automate tasks like splitting bills in Cash App or managing business finances
• The agent shows promise in creative pursuits, including music generation
• Future plans include potential on-device deployment for mobile applications

Strategic context; Block’s release of goose aligns with their broader commitment to open-source development and comes amid significant industry developments.

• The launch coincides with Block’s establishment of an Open Source Program Office
• Goose enters the market alongside other recent releases like DeepSeek’s R1 and new AI assistants from Perplexity and OpenAI
• The agent’s flexibility and open-source nature position it uniquely in the growing AI assistant marketplace

Looking ahead: Market implications; While goose currently excels in development tasks, its architecture and adaptability suggest potential for broader impact across various sectors, though success will likely depend on community adoption and continued development of practical use cases beyond coding.