About this AI skill
The Feature prototype skill is pre-built for Elle (the AI assistant in Aha! software) and creates an interactive prototype of a feature. Elle reads the feature's description, requirements, and linked ideas. It then generates the underlying code and a preview that covers the primary workflow, an empty state, and an error state using realistic sample data. The output also notes which requirements the prototype shows, what it leaves out, and any assumptions made, along with suggested questions to validate.
The full Feature prototype skill, as Elle uses it:
# Feature prototype
Use once a feature has enough definition to visualize, to test the concept and gather feedback before development starts.
## Definitions
Use these meanings consistently in every step.
- **Feature:** The feature record being prototyped, or the feature whose record link the user pasted into the chat
- **Fidelity:** How finished the prototype looks. Low shows structure and flow only. Medium adds real labels and states. High applies visual design.
- **Scope:** The specific screens and states the prototype covers. Everything outside that set is out of scope.
- **State:** A condition a screen can be in, such as default, empty, loading, error, or a permission-restricted view
## Workflow
1. Read the feature, its description, its requirements, and its linked ideas. The requirements define what the prototype has to demonstrate.
2. Confirm fidelity and scope with the user. Default to medium fidelity and the primary workflow only.
3. Identify the workflow to show, from entry point to completion. One complete path beats fragments of several.
4. Generate realistic sample data drawn from the feature and the workspace. Use plausible names, dates, and values.
5. Build the prototype. Cover the primary path first, then the empty state and one error state.
6. Note which requirements the prototype demonstrates and which it does not, so reviewers know what they are looking at
7. List what the prototype deliberately leaves out, so a gap is not read as a design decision
8. Suggest what to validate and with whom, drawing on the organizations behind any linked ideas
## Rules
- If the feature has no requirements, say so and prototype the primary path only, marking the assumptions made
- Match the visual conventions of Aha! software unless the user supplies a different design reference
- When the user uploads an image as a design reference, follow its layout and styling
- Sample data is realistic and internally consistent. Never leave lorem ipsum in a prototype.
- Reference the feature and its requirements by reference number and name
- Do not update the feature. Present the prototype and let the user decide.
- Headers use sentence case. No em dashes, en dashes, or curly quotes.
## Output
```
## What this prototype shows
<the workflow it covers, in two to three sentences>
## Screens
| Screen | Purpose | States covered |
## Requirements demonstrated
| Requirement | Shown |
## Out of scope
- <what was deliberately left out>
## Assumptions made
- <where the record was silent and a choice was made>
## What to validate
- <question, and who to ask>
```
Save the prototype as an artifact and share the link so the team reviews the same version.
## Quality checks
- The prototype covers one complete path rather than several partial ones
- Every requirement is marked as shown or not shown
- Sample data is plausible and consistent across screens
- Assumptions are listed rather than embedded silently in the designTry this AI skill now
Open the Feature prototype skill in Aha! Roadmaps to build an interactive prototype of a feature.
Open this AI skillAbout Elle
Elle , the AI assistant in Aha! software, helps product teams accelerate product work by creating content, analyzing records, and updating plans using information in their account. The AI prompt library offers pre-built skills and tasks for Elle that are tuned for product development work.