@ -10,6 +10,8 @@ At the center is **ABP Agent**, our AI coding assistant. Around it are long-stan
## Meet ABP Agent
ABP Agent is the AI coding assistant built into ABP Studio. It operates in three modes, each tuned for a different stage of work:
- **Agent mode**: the implementation mode. The agent reads the solution, writes and edits files, builds the affected projects, runs your apps, watches the runtime, and iterates until the change works end-to-end.
@ -18,8 +20,6 @@ ABP Agent is the AI coding assistant built into ABP Studio. It operates in three
The agent is **ABP-aware by default**. It is instructed to prefer ABP base classes over plain POCOs, repositories over direct `DbContext` injection, `ApplicationService` over plain services, the ABP permission system over `[Authorize(Roles=…)]`, localized strings over hardcoded text, `BusinessException`/`UserFriendlyException` over plain `Exception`, the distributed cache abstraction over raw memory cache, and background job abstractions over hand-rolled hosted services. When it is unsure about an ABP feature, it consults the official ABP documentation as a primary source of truth, not random blog posts on the web.
---
## Why It Was Needed
@ -41,6 +41,8 @@ ABP Agent and the surrounding ABP Studio features were built to close exactly th
## How ABP Agent Sees Your Application: The Analyze Engine
Before ABP Agent answers anything, it needs to *understand* your solution. This is where the **Analyze** feature does the heavy lifting.
When you open a solution, ABP Studio analyzes every package and builds a structural map: the application's **skeleton**. It identifies what each type actually is in ABP terms: an aggregate root, an entity, a value object, a repository interface or implementation, a domain service, an application service or interface, a DTO, an integration service, a controller, an HTTP API, a background job or worker, an event handler, an ETO, a SignalR hub, a DbContext (EF Core or MongoDB), a permission/feature/setting provider, a global feature, a Mapperly or AutoMapper profile, a fluent validator, a menu contributor, a data seed contributor, an options class, an ABP module with its `[DependsOn]` chain, and many more.
@ -126,6 +128,8 @@ The Task Runner integration is what makes the *After* phase especially valuable.
## Git & GitHub: Reviewing and Committing Without Leaving the IDE
ABP Studio now ships a full Git client with deep GitHub integration. The goal is simple: once the agent finishes a change, you should be able to review, commit, push, branch, and respond to review feedback **without ever leaving ABP Studio**.
The Git side covers everything you'd expect:
@ -151,8 +155,6 @@ Two AI-assisted touches make this loop especially smooth:
The result is a *closed loop*: the agent writes the change, you (or the AI reviewer agent) review the diff, you let the agent fix the comments, you commit with an AI-suggested message, you push, and you head to GitHub for the pull request, all from inside ABP Studio.
İsmail ÇAĞDAŞ
2 weeks ago
