Iterate on tasks

Iterations are a core concept in Rover. You might be happy with the first implementation of a task made by the agent, but in most situations, you’ll want to refine what it did. This guide shows you how to iterate on tasks to get the results you need.

Understanding iterations

You can have as many iterations as you need as part of a task. They reflect the interaction between you and the agent with regards to a specific task. Whenever you ask Rover to create a new iteration on a task, all previous iterations are taken into account and become part of the context of the overall task, leading to the final result—the last iteration.

Create an iteration

Following the fizzbuzz example from the task creation guide, you can ask for an iteration to refine the implementation.

Task Iteration

1
Request an iteration to modify the task. For example, update all output to use FIGlet style.
Terminal window
```
rover iterate 1 "Update all output to use the figlet style"
```
2
As with task creation, you can follow the logs of the iteration as it happens. The task number does not change with new iterations.
Terminal window
```
rover logs -f 1
```

Review iteration changes

After an iteration completes, you can inspect what changed compared to the previous iteration.

Check iteration diff

1
View the diff to see what the agent modified
Terminal window
```
rover diff 1
```
2
Inspect the iteration details to see the agent’s plan and summary
Terminal window
```
rover inspect 1
```

When to iterate

Iterations are useful when you want to:

Refine the implementation: Ask the agent to improve code quality, add error handling, or optimize performance
Add new features: Extend the existing implementation with additional functionality
Fix issues: Address bugs or problems discovered during testing
Adjust style: Change coding style, naming conventions, or formatting
Update documentation: Improve comments, READMEs, or inline documentation

Best practices

To get the best results from iterations:

Be specific: Provide clear, actionable feedback about what needs to change
One focus per iteration: Keep each iteration focused on a specific improvement
Test between iterations: Verify changes work as expected before requesting the next iteration
Use context: Reference specific files, functions, or line numbers when possible
Build incrementally: Start with core functionality and add complexity through iterations

Relevant Concepts

Task Learn about Rover tasks and how they work

Workflow Discover how workflows guide coding agents

Implement a new feature Create and assign tasks to AI coding agents

Merge changes from tasks Integrate completed tasks into your codebase

Project Initialization Initialize Rover in your repository

Write technical documentation Use Rover to create and maintain documentation