> 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/sonarqube-server/user-guide/issues/with-ai-features/backlog-fix-suggestions.md).

# Backlog fix suggestions

Request and review backlog fixes with the Remediation Agent in SonarQube Server 2026.5.

Use the SonarQube Remediation Agent to request fixes for static analysis issues and Hunter Agent findings in your backlog. The agent generates changes in the background and opens pull requests in your repository for you to review and approve.

The Remediation Agent is available in the Enterprise and Data Center editions and requires a separate subscription.

## Prerequisites <a href="#prerequisites" id="prerequisites"></a>

Before requesting fixes manually:

* An instance administrator has deployed and enabled the Remediation Agent, configured an LLM provider and model, and turned on **Manual backlog remediation**. See [Set up the Remediation Agent](/sonarqube-server/instance-administration/ai-features/remediation-agent.md#setup-overview).
* Your project is bound to GitHub, GitLab, or Azure DevOps, and the agent has write access to the repository. See [DevOps platform permissions](/sonarqube-server/instance-administration/ai-features/remediation-agent.md#devops-platform-permissions).
* You're signed in and have access to the project. No dedicated permission is required to request a fix.

## Eligible issues <a href="#eligible-issues" id="eligible-issues"></a>

Issues must have an **Open**, **Confirmed**, or **Accepted** status:

* Static analysis issues must be in C#, Java, JavaScript, TypeScript, or Python. Some rules are [unsupported](/sonarqube-server/instance-administration/ai-features/remediation-agent.md#unsupported-rules).
* Hunter Agent findings are eligible regardless of language. To generate findings, your instance also needs a separate [Hunter Agent subscription and setup](/sonarqube-server/instance-administration/ai-features/hunter-agent.md).

## Assigning issues to the agent <a href="#assigning-issues" id="assigning-issues"></a>

To request a fix for one issue, open your project's **Issues** page and select **Fix with agent** in the issue row or on the issue's details page. The button appears only when the issue is eligible and the agent is available for your project.

To request fixes for several issues:

1. Go to your project's **Issues** page.
2. Select the issues you want fixed.
3. Select **Fix \<count> with agent**.
4. Open **Project** > **Remediation Agent** in the project navigation to follow the jobs and review the linked pull requests.

You can assign up to 20 issues at a time. If you select more, only the first 20 eligible issues in list order are sent. Ineligible issues are skipped.

If only some assignments succeed, a warning appears and the issues that still need attention remain selected. Retry the failed assignments.

When automated backlog remediation is enabled, the agent also proposes fixes on a schedule, without anyone assigning issues. See [Remediation Agent (Project administration)](/sonarqube-server/project-administration/ai-features/remediation-agent.md).

### Fixing Hunter Agent findings <a href="#fixing-hunter-agent-findings" id="fixing-hunter-agent-findings"></a>

Filter the **Issues** page by the `hunter-agent` tag to find [Hunter Agent findings](/sonarqube-server/user-guide/issues/with-ai-features/hunter-agent-findings.md). Request fixes individually or select several findings using the same steps as for other issues.

Each finding starts a separate job and produces its own pull or merge request if a fix is available.

## Tracking the agent's jobs <a href="#tracking-the-agents-jobs" id="tracking-the-agents-jobs"></a>

To follow what the agent is doing, open **Project** > **Remediation Agent** in the project navigation. It lists each job with:

| Column                | Description                                                                                                                                                                                                                                                                                                                                          |
| --------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Status / Duration** | The job's current state: **Pending**, **In progress**, **Completed**, or **Failed**. Completed and failed jobs also show how long the job took. An in-progress job shows **Please wait...** instead.                                                                                                                                                 |
| **Started**           | When the job started. Recent jobs show **Today** or **Yesterday**.                                                                                                                                                                                                                                                                                   |
| **Source**            | What the job came from. **Manual** means someone assigned static analysis issues or a Hunter Agent finding. **Automated** means a scheduled run. **Pull request** means the job was triggered from a failing pull request; see [Pull request fix suggestions](/sonarqube-server/user-guide/issues/with-ai-features/pull-request-fix-suggestions.md). |
| **Outcome**           | A link to the pull request the agent opened, shown as **PR #\<number>**. Pending and in-progress jobs show a dash, as do completed jobs that did not produce a pull request. A failed job shows **Failed** with a tooltip giving the reason.                                                                                                         |

The page shows the number of jobs currently listed. Select **Show more** to load older jobs.

If the project has no jobs yet, the page shows **No agent jobs yet**. If the list cannot load, the page shows **Something went wrong while loading agent jobs.** with a **Retry** action.

## Reviewing the agent's pull request <a href="#reviewing-the-pull-request" id="reviewing-the-pull-request"></a>

The agent opens a pull request in the bound repository containing its proposed fixes. Review it the way you review any other pull request in your team's workflow, then merge it to apply the fixes.

Fixed static analysis issues close on the next analysis of the target branch. Hunter Agent findings close when the next Hunter Agent analysis of that branch no longer detects them.

### Reviewer assignment <a href="#reviewer-assignment" id="reviewer-assignment"></a>

The agent automatically assigns a reviewer to the pull request it opens, using each DevOps platform's native reviewer concept: GitHub reviewers, GitLab approvers, or Azure DevOps required reviewers.

To choose a reviewer, the agent looks for someone who recently worked on the changed files. If a file has no history, it falls back to whoever committed recently elsewhere in the repository.

Reviewer assignment does not block the pull request: if the agent cannot confirm an eligible reviewer, it still opens the pull request without one.

On forked repositories, the agent's candidates can include the upstream project's authors, who are not collaborators on your fork. In that case, reviewer assignment can fail and the pull request opens without a reviewer.

On GitLab, reviewer assignment requires GitLab 16.3 or later. See [Remediation Agent](/sonarqube-server/instance-administration/ai-features/remediation-agent.md#reviewer-assignment-support).

## Related pages <a href="#related-pages" id="related-pages"></a>

* [Remediation Agent](/agent-centric-development-cycle/in-your-long-living-branches-the-code-maintenance-loop/remediation-agent.md)
* [Get backlog fix suggestions](/agent-centric-development-cycle/in-your-long-living-branches-the-code-maintenance-loop/solve-issues/backlog-fix-suggestions.md)
* [Remediation Agent](/sonarqube-server/instance-administration/ai-features/remediation-agent.md)
* [Remediation Agent](/sonarqube-server/project-administration/ai-features/remediation-agent.md)
* [Hunter Agent findings](/sonarqube-server/user-guide/issues/with-ai-features/hunter-agent-findings.md)


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.sonarsource.com/sonarqube-server/user-guide/issues/with-ai-features/backlog-fix-suggestions.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.