Aspire for AI Applications
summary
Aspire adds an MCP server now runnable from the CLI so coding agents get full app context (logs, traces, config) and can execute commands like restart or restore; demo turns a mock weather app into a real one by calling an Open‑Meteo free weather API library, edits backend and frontend, and uses Playwright to let an agent "eyes" via Playwright verify UI interactions and traces. Aspire 13.1 exposes MCP to VS Code, shows a GenAI Visualizer for model conversations, recovers from Docker hiccups, and the agent iteratively fixes and validates the changes.
AI-written summary · about
chapters (15)
- 0:00Welcome to AI Dev Days: Introducing Aspire
- 0:23What is Aspire? Overview of Features and Dashboard
- 1:10From Assistants to Agents: The Evolution of AI in Development
- 2:34Introducing MCP in Aspire: Context for Coding Agents
- 3:45Demo Setup: Aspire MCP and CLI Enhancements
- 5:11Configuring MCP for VS Code and Agent Integration
- 7:04Live Demo: Querying Aspire Resources via MCP
- 9:01Giving Agents “Eyes”: Logs, Traces, and Front-End Control
- 12:02Adding Real Weather API: Agent-Driven Code Updates
- 15:00Enhancing the UI: Dynamic City Selection and Testing
- 18:02Closing the Agent Loop: Self-Validation and Orchestration
- 19:27Second Demo: Adding AI to Aspire Bingo App
- 21:04Building a Harness for AI Integration and Classification
- 22:03Using Aspire Dashboard and GenAI Visualizer for Insights
- 24:34Recap: Assistants vs Agents, Aspire’s Role, and Next Steps
related talks
01h 00mBuilding next-level conversational AI apps with .NET and Aspire
44m.NET Aspire development on any OS with the Visual Studio family
01h 08mCode to cloud with .NET Aspire
37mLetting AI Interface with Your App with MCPs
28mBuilding MCP servers with VS Code — Level up your MCP
32mProgramming language evolution in the AI era
01h 28mAI Coding Agents and how to code them
29mStop Rewriting. Start