PrisAImSearch
REAL IMPLEMENTATIONS / EVERYDAY

Breaking large tasks into steps with Goblin Tools

Bram De Buyser built Magic ToDo to turn a large task into smaller actions you can edit and check off.

THE ACTUAL IMPLEMENTATION

What they built.

Bram De Buyser built Goblin Tools around a familiar problem: knowing what needs to be done is not the same as knowing where to start. Magic ToDo takes a task written in ordinary language and proposes smaller actions. The useful output is an editable checklist that a person can reshape, rather than an agent silently carrying out the work.

The live web tool and its creator documentation establish that this interface was implemented and made available. The current About page describes general-purpose models from multiple providers without naming exact versions. The example therefore explains the product interaction without attributing it to an assumed GPT or Claude model.

Shipped tool

The public interface and creator documentation establish the implementation. They do not establish clinical benefit or measured time savings.

Try Magic ToDo

How the tools work together

The person supplies the task and chooses the amount of detail through the spiciness control. The AI proposes subtasks, and the person edits, reorders or completes them. That review is part of the design: a useful breakdown depends on the person's situation, so a generated checklist should not be treated as a complete plan for everyone.

Workflow at a glance

  1. Write the task

    Add the thing you need to do to Magic ToDo.

  2. Choose the amount of detail

    Adjust the spiciness setting, then generate subtasks.

  3. Make the list your own

    Edit, reorder and check off useful suggestions. Remove steps that do not fit your situation.

How you could adapt the idea

To adapt the idea, begin with one ordinary task whose result you understand, such as organizing a desk or preparing a simple packing list. Compare the suggested actions with the real situation and remove assumptions that do not apply. This is an editorial way to evaluate the pattern, not a clinical claim or a measured saving. The core web tools can be tried without building your own agent; Pro features and mobile apps have separate terms.

Effort, cost and limitations

Effort
The core web tool can be tried without building a workflow.
Cost
Core web tools are free. Advanced Pro features and paid mobile apps are separate.

Suggestions can be incomplete or inappropriate; adjust them to the real task.

Go to the original work.

Magic ToDo and its helpCreator, models, costs and limitations

Source-backed editorial account; PrisAIm did not run or independently benchmark this implementation.

Start here: AI basicsNewsCompareWorkshopsDiscovery methodsSearch everything
Browser-local saved list

Clear your saved entries?

This removes your list from this browser only. You can save entries again at any time.