> For the complete documentation index, see [llms.txt](https://docs.sonarsource.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.sonarsource.com/agent-centric-development-cycle/readme.md).

# About the Agent Centric Development Cycle

## Overview

The Agent Centric Development Cycle (AC/DC) is Sonar’s framework for building software in an AI-driven world, where agents handle much of the coding and humans govern quality.\
It defines a continuous 3-stage loop for every AI-assisted change:

**Guide → Verify → Solve**

* **Guide** provides project context to agents before they write or edit code.
* **Verify** analyzes the resulting code.
* **Solve** fixes issues found during verification.

This workflow helps agents produce code that aligns more closely with project standards, catches issues earlier, and shortens the review-to-fix cycle.

The diagram below shows where AC/DC features fit in this cycle.

```mermaid
%%{init: {'theme': 'base', 'themeVariables': {
  'primaryColor': '#E6F2FF',
  'primaryBorderColor': '#126ED3',
  'primaryTextColor': '#111827',
  'clusterBkg': 'rgba(18,110,211,0.06)',
  'clusterBorder': '#126ED3',
  'lineColor': '#126ED3'
}}}%%
flowchart LR
  %% Phases laid out left-to-right, details descending within each phase

  subgraph GuidePhase["Guide"]
    direction TB
    CA["Context Augmentation
---
Project-specific context before work starts"]
  end

  subgraph VerifyPhase["Verify"]
    direction TB
    AA["Agentic Analysis
---
Analyzes code with full CI context"]
    ACR["Gitar
---
Inline comments, summaries,
walkthroughs, diagrams on PRs"]
  end

  subgraph SolvePhase["Solve"]
    direction TB
    RA["Remediation Agent
---
Fixes PR & backlog issues,
re-analyzes fixes"]
    CF["AI CodeFix
---
Generates fixes for individual SonarQube issues"]
  end

  %% Left-to-right flow between phases
  CA --> AA
  CA --> ACR
  AA --> RA
  AA --> CF
  ACR --> RA

  %% Re-verify after fixes
  RA --> AA
  CF --> AA
```

## Guide

Context Augmentation provides project-specific context (coding guidelines, architectural information, and codebase context) to the AI agent before it writes or edits code. This reduces avoidable violations and inconsistent code patterns.

## Verify

Agentic Analysis analyzes code during the agent workflow. The agent can use the analysis results to refine the code before review. This surfaces quality and security issues before they reach pull request review or CI.

[Gitar](https://gitar.ai), a separate Sonar product, posts inline review comments, change summaries, and on-demand walkthroughs and architecture diagrams directly on pull requests.

## Solve

The Remediation Agent generates fix suggestions for issues found in pull requests and in the backlog. AI CodeFix generates fix suggestions for individual issues detected by SonarQube. This reduces manual rework and supports incremental cleanup of existing issues.

## Get started

See the [feature pages](/agent-centric-development-cycle/features/context-augmentation.md) for product details and the [how-to guides](/agent-centric-development-cycle/how-to-guides/add-context-to-generate-better-code.md) for setup and usage.

## Related pages

* [Code quality and security in the age of AI and agentic development](https://www.sonarsource.com/resources/library/buyers-guide-to-code-quality-and-security/)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.sonarsource.com/agent-centric-development-cycle/readme.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.