liveBook · Manning

AI agent

As explained in Building Distributed Applications that Work

An AI agent uses artificial intelligence to perceive context, make decisions, or perform tasks.

An AI agent can assist with application work or power an AI application. In an Aspire workflow, a coding agent inspects a .NET solution, proposes an AppHost model and implementation plan, and helps make changes incrementally. Semantic Kernel is identified as a framework for building .NET agents that interact with large language models. section 10.4.2 section 5.4

Ask for a plan before changing the solution

A useful prompt is: “Inspect this solution and propose an Aspire AppHost model (projects+dependencies) and the order to implement it.” Review the proposal, then make changes in small iterations and verify them before shipping. section 10.4.2

Listing 10.7 Claude output describing the solution structure
Claude: I'll scan the repo structure, solution/projects, and configuration files (appsettings, launchSettings, docker-compose) to infer runtime dependencies and propose an Aspire AppHost model + an incremental implementation plan.

🔎 Searching for .sln files...
- Found: OnlineShop.slnx

🔎 Parsing solution projects...
- OnlineShop.ApiService/OnlineShop.ApiService.csproj
- OnlineShop.Web/OnlineShop.Web.csproj

🔎 Checking configuration and local orchestration files...
- Found: OnlineShop.ApiService/appsettings.json
- Found: OnlineShop.Web/appsettings.json
- Found: OnlineShop.ApiService/appsettings.Development.json
- Found: OnlineShop.Web/appsettings.Development.json
- Found: OnlineShop.ApiService/Properties/launchSettings.json
- Found: OnlineShop.Web/Properties/launchSettings.json

🔎 Inferring dependencies from config...
- Database: SQL Server (ConnectionStrings:DefaultConnection -> Server=(localdb)\\mssqllocaldb;Database=OnlineShopDb;...)

🔎 Inferring service-to-service calls...
- OnlineShop.Web calls OnlineShop.ApiService:
  - Found HttpClient named "ProductsApiClient" base URL: https://localhost:7465 (appsettings.Development)

✅ Proposed Aspire AppHost model

Projects:
1) web     -> OnlineShop.Web            (external http endpoint)
2) api     -> OnlineShop.ApiService     (internal http endpoint)

See an example of a coding agent inspecting an online-shop solution’s projects and configuration to infer dependencies. section 10.4.2

How Building Distributed Applications that Work covers it

How deep it goes
Works through codebase analysis, AppHost proposals, implementation plans, and verification, but does not define agent internals.
Examples
Demonstrates Claude Code inspecting an online-shop solution, sketching its AppHost, and proposing a staged implementation plan.
What it assumes
Assumes access to a .NET solution with projects to integrate into Aspire and recommends iterative review for less-experienced users.
What you can do after
Apply a coding agent to analyze a solution, propose Aspire changes, implement them incrementally, and verify the result before shipping.
How it connects
Links AI agents to large language models, Semantic Kernel, .NET, and Aspire AppHost orchestration.

Also explained in

Compare how 33 books explain AI agent →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Building Distributed Applications that Work. The section links point to the parts of the book it draws on.