Goose AI agent and MCPUI demo for GitHub issue management
A GitHub Universe 2025 demo shows how the Goose AI agent paired with the GitHub MCP server can create and assign GitHub issues from the terminal and render interactive visual interfaces via MCPUI.
Score breakdown
Developers building agentic workflows can use the Goose + GitHub MCP server combination to automate issue management from the terminal, while MCPUI opens the door to agents that return interactive visual outputs rather than plain text responses.
- 01Demo was presented at GitHub Universe 2025.
- 02The Goose AI agent is used together with the GitHub MCP server.
- 03The demo shows automatic creation and assignment of a GitHub issue from the terminal.
Presented at GitHub Universe 2025, this demo explores the integration of the Goose AI agent with the GitHub MCP server for terminal-based GitHub issue management. The demonstration covers automatically creating and assigning a new GitHub issue without leaving the terminal, illustrating how agentic workflows can streamline common developer tasks.
The demo also introduces MCPUI, which enables AI agents to respond with fully interactive visual interfaces rather than plain text.
The demo also introduces MCPUI, which enables AI agents to respond with fully interactive visual interfaces rather than plain text. As an example, the agent renders a team project calendar, suggesting that MCPUI can surface structured, graphical outputs directly within an agentic workflow. Together, Goose and MCPUI point toward a model where AI agents handle both backend actions and rich front-end presentation.
Key facts
- 01Demo was presented at GitHub Universe 2025.
- 02The Goose AI agent is used together with the GitHub MCP server.
- 03The demo shows automatic creation and assignment of a GitHub issue from the terminal.
- 04MCPUI allows AI agents to respond with fully interactive visual interfaces.
- 05A team project calendar is shown as an example of an MCPUI-rendered interface.
Topics
Summary and scoring are generated automatically from the original article. We always link back to the publisher and never republish images or paywalled content. Last processed Apr 24, 2026 · 17:11 UTC. How this works →