Use cases
This section is a collection of practical recipes: each page starts from a concrete goal, walks through the numbered steps, and ends with a verifiable result — a generated file, a published skill, a scheduled routine, or a completed code review.
These are "how-to" guides. For the mental model behind each feature, follow the links to Core concepts and to the per-surface guides under Using Imaginne.
Where to start
If you haven't installed or signed in yet, start with the Quickstart. Then pick the recipe closest to what you need to do right now.
The recipes
Build a .pptx deck by asking in plain language — the agent uses the native ppt_skill, which picks the visual on its own.
Point to a file with @data.csv, ask for the summary, and get a ready-made .xlsx.
From Skill Studio to publishing: turn a process from your company into a reusable skill.
See recipe →Automate a weekly reportSchedule a Routine in the Desktop app, get the notification, and find the artifacts in the project.
See recipe →Review code in VS CodeReview your changes, apply fixes, and generate the commit message without leaving the editor.
See recipe →Beyond these recipes, the agent also generates images from prompts: describe the image you need and get an image artifact ready to use.
How to read a recipe
Every page follows the same structure, so you can execute fast:
| Section | What it covers |
|---|---|
| Goal | What you'll have at the end. |
| Prerequisites | The login, surface, and permissions you need. |
| Steps | The numbered sequence, with the actual commands and prompts. |
| Result | The expected artifact or effect, and where to find it. |
Ask for something concrete and verifiable in your first request (e.g., "summarize this CSV in 3 highlights and generate an .xlsx"). It's easier to review the result and adjust than to ask for everything at once.
See also
Was this page helpful?
Report a problem on this pageDo not send passwords, keys, tokens, or customer data.