How to Write Effective AI Prompts: Best Practices and Practical Examples

AI Apps Builder turns your plain-English description into working Forge apps: global pages, dashboards, gadgets, or issue panels. The quality of the generated app depends directly on the clarity and structure of your prompt.

This guide combines:

• AI prompt engineering best practices.

• Real use cases from AI Apps Builder.

• Patterns from successful dashboard, gadget, issue panel, and app prompts.

1. Start Your Prompt With a Clear Goal

According to prompt engineering best practices, an effective prompt begins with a clearly defined objective. For AI Apps Builder, this means:

✔ Define what you want to build
✔ Define where it should work in Jira
✔ Define the business outcome

Basic prompt:

Create a dashboard gadget that tracks sprint health and progress.

Improved prompt:

Create a dashboard gadget that tracks sprint health and progress for a selected project and sprint. Include a burndown chart and a summary of completion status. Use Jira style for UI.

To ensure accurate results, clearly state your goal. AI Apps Builder performs best when a defined goal is set.

2. Define the App Type First

AI Apps Builder can generate different types of Jira apps, including dashboards, gadgets, issue panels, and full-page apps. Each of these modules works in a different place inside Jira and follows different structural rules. For that reason, always start your prompt by clearly stating what you want to build.

Examples of module types:

• Gadget or widget for dashboard — appears in the list of gadgets on Jira dashboards.

• Issue Panel — appears inside a Jira issue.

• Global Page (app or dashboard) — accessible from the Apps menu.

Example

Instead of:

Show sprint progress.

Write:

Create an Issue Panel for Jira that shows an overview of the current sprint.

Clearly naming the module helps the AI build the right structure from the start.

3. Be Explicit About Data Scope

AI Apps Builder uses the information you explicitly provide in your prompt. You need to specify:

• Which project you mean

• Which sprint you care about

• Which statuses define “cycle time”

• Which issues to include or exclude

• Over what period should metrics be calculated

Examples: What You Should Define:

Time range

• Last 3 sprints

• Selected date range

• Current sprint

• Past 30 days

Status boundaries

• From “In Progress” to “Done”

• Only unresolved issues

• Exclude closed sprints

Project scope

• Selected project

• Multiple selected projects

• All company projects

Issue types

• Stories only

• Bugs and tasks

• Exclude subtasks

Assignee rules

• Group by assignee

• Only active users

• Only users who logged time.

Weak prompt:

Create a gadget that shows sprint velocity.

Stronger prompt:

Create a gadget that shows velocity for the last 3 sprints, plotting committed vs completed story points and calculating average velocity.

4. Define Business Logic Clearly

When logic is clearly defined, the AI can generate precise and predictable results. For instance, define high-risk as: flagged overdue issues or issues due within 7 days.

For calculated fields, specify:

• What to calculate

• From which statuses

• Over what period

• What comparison to show

5. Specify Visual Structure

If you describe the layout, format, or structure clearly, the AI Apps Builder is more likely to generate what you expect than if you keep it abstract.

Define:

• Charts (burndown, trend line, bar chart)

• Cards

• Tables

• Columns

• Color logic

• Status indicators.

6. Define UI Expectations

AI Apps Builder generates both logic and interface (Custom UI). If you don’t describe how the app should look, the layout may be technically correct, but visually inconsistent with your expectations.

You can define:

• Use Jira style for UI

• Follow Atlassian Design System principles

• Use tabs to separate sections

• Use badges for status indicators

• Use color-coded labels.

Example

Basic:

Create a sprint dashboard with completion metrics.

Improved:

Create a sprint dashboard with completion metrics. Use Jira style for UI, display summary metrics as cards at the top, and use a clean two-column layout.

7. Separate Summary and Detailed Views

Try to think like a product designer and ask yourself:

• What should users see immediately?

• What should they explore next?

• What details should appear only after interaction?

When you define these layers in your prompt, AI Apps Builder generates more structured and professional results.

Example

If you only say:

Show team workload.

The result may lack hierarchy or clear structure.

If you say:

Show total hours per user as summary cards, include a bar chart for comparison, and allow clicking a user to open a detailed issue table.

Now the system clearly understands:

• What to show first (summary cards)

• What to visualize (bar chart comparison)

• What to reveal on demand (detailed issue table)

Clearly defining how users interact with the app improves the quality and usability of the generated app.

8. Define Interaction Elements

When writing your prompt, clearly describe how users should interact with the app. Include elements such as:

• Dropdown filters

• Multi-select filters

• Date range pickers

• Click behaviors

• Expandable sections

Example:

If you write:

Show workload per user.

The result may display one general view without filtering options.

If you write:

Add multi-select filters for project and assignee.

Now the app becomes flexible. Users can adjust the view without rebuilding the tool.

Recommended Prompt Structure for AI Apps Builder

You can use this template:

Create a [dashboard gadget/issue panel/widget/app] that [main purpose].

Data Scope:

• [projects]

• [time range]

• [statuses]

• [issue types]

Display:

• [charts]

• [summary metrics]

• [tables with columns]

Filters:

• [dropdowns / multi-select / date range]

Logic:

• [definitions of conditions]

Visual Indicators:

• [color rules]

Interactions:

• [click actions / drill-down]

UI:

• [Jira style / Atlassian design]

• [layout preference].

Example (Structured Prompt)

Create a dashboard gadget that analyzes sprint performance.

Data Scope:

• Selected project

• Last 3 completed sprints

Display:

• Velocity trend chart (committed vs completed story points)

• Average velocity

• Percentage change compared to the previous sprint

Filters:

• Project dropdown

Visual Indicators:

• Upward trend in green

• Downward trend in red

Interactions:

• Click the sprint to see the detailed issue list.

UI:

Use Jira style for UI.

Short Checklist Before You Click “Generate”

Use this quick checklist:

✅ Did I clearly state what I want to build?
✅ Did I define the Jira module type?
✅ Did I specify time range, scope, and statuses?
✅ Did I define any business logic explicitly?
✅ Did I specify charts, tables, and layout?
✅ Did I define filters and interactions?
✅ Did I include visual indicators (if relevant)?
✅ If something is unknown, did I specify assumptions?

If you can answer “yes” to most of these, your prompt is strong.

Haven't used AI Apps Builder yet? You're welcome to try it then!

Have questions or want to see AI Apps Builder in action?
Feel free to meet with us — book a demo today!