Remember that saying, "it’s not what you ask, but how you ask"?

When I first started working with goose as an AI agent, I was convinced there had to be one ‘best' prompting style. I spent so much time trying to figure out which one was superior, but the more I used goose, the more I realized that couldn't be further from the truth. There isn’t one right way to prompt AI, but there are better approaches depending on what your end goal is.

So, let’s go through which prompt style works best for your specific needs, and how you can use them to vibe code a little better with goose.

How do we debug AI agents like goose? In the Goosing Around stream series, host Rizel Scarlett invited Marc Klingen, Co-Founder at Langfuse, and Alice Hau, Machine Learning Engineer at Block, to demo how Langfuse enables observability into goose's actions, letting you trace LLM behavior and catch errors.

On the Goosing Around stream series, host Rizel Scarlett demonstrated how to use goose locally with Ollama for a fully local experience on your device. Her guest, Parth Sareen, an experienced software engineer with a focus on building frameworks and systems for AI/ML, showed us the magic of structured outputs and how goose and Ollama work together under the hood.

In our previous episode of goose Flight School, our host Adewale Abati showed us how to use goose to take a Figma design and transform it into a functional Nuxt application. In this stream he covered the entire process, from initial setup to final implementation, highlighting how goose can help developers bridge the gap between design and development.

In the latest episode of Wild goose Case, hosts Ebony Louis and Ace Abati explored a fascinating new way to extend goose’s automation capabilities by integrating with VOYP, an AI-powered system that makes phone calls. Their guest, Paulo Taylor, a technology veteran with over 35 years of experience, walked through how developers can use goose to trigger and manage phone-based interactions through VOYP.

Working with AI agents can sometimes feel unpredictable. After using goose extensively for the last few months, I've compiled a few key tips that will help you get the most out of this tool. No matter your workflow, these guidelines will help you work more efficiently with goose.

During our previous livestream, Aaron Goldsmith, Infrastructure Operations Engineer at Cash App, showed a team of goose AI agents collaborating in real time to create a website. Our community loved it so much, Cliff Hall was inspired to iterate on that idea and create a gooseTeam MCP server.

It seems like yesterday when we all were wowed by generative AI and specifically the chat interfaces that made interacting with large language models (LLMs) accessible to everyday people.

As amazing as this was, it was only the beginning. The next wave of AI is agentic, meaning AI systems that don't just respond to prompts but take actions, make decisions, and interact with external systems. This is accomplished via AI agents.

We are thrilled to announce codename goose, your on-machine, open source AI agent built to automate your tasks.

Powered by your choice of large language models (LLMs), a user-friendly desktop interface and CLI, and extensions that integrate with your existing tools and applications, goose is designed to enhance your productivity and workflow.

Running into Continuous Integration (CI) failures in pull requests can be quite frustrating but they happen very often. In this post, we leverage the GitHub CLI (gh) using goose to simplify your CI debugging process, fetch detailed information about failed CI runs and annotations directly from GitHub, and even apply fixes directly.