Using Ascona AI inside Jira

1. What Ascona AI does in Jira

Ascona AI is an AI project management copilot that lives inside Jira. It takes your high-level product idea or feature description and turns it into a structured, actionable backlog:

• Generates Epics, Stories, and Tasks directly into your Jira project.
• Writes ticket descriptions, acceptance criteria, and checklists.
• Suggests estimates and priorities so you can plan sprints faster.
• Keeps everything editable – you stay in control, AI just does the heavy lifting.

2. Prerequisites

Before using Ascona AI inside Jira, make sure you have:

• A Jira Cloud instance with permission to install Marketplace apps (site or org admin).
• An active Ascona AI account (or the one used in the demo) with Jira integration enabled.
• At least one Jira software project where you want Ascona AI to create issues.

Tip: Start by connecting Ascona AI to a sandbox or test project first. Once you like the structure and ticket quality, switch to production projects.

3. Installing the Ascona AI app in Jira

3.1. Install from Atlassian Marketplace

1. In Jira, click the Apps menu in the top navigation bar.
2. Select Explore more apps (or Find new apps).
3. In the Marketplace search bar, type Ascona AI.
4. Open the Ascona AI app listing and click Try it free or Install.
5. Confirm the permissions and wait for the installation to complete.

3.2. Granting permissions

On first install, Jira will ask you to grant the app access to projects and issues. Ascona AI needs this to:

• Create new issues (Epics, Stories, Tasks, Sub-tasks).
• Read existing issues to avoid duplicates and to align with your structure.
• Update fields like description, acceptance criteria, labels, and story points.

Security note: Only grant access to projects where you actually want Ascona AI to work. You can restrict the app later if needed using Jira’s permission settings.

4. Connecting Ascona AI to your Jira workspace

4.1. First-time setup

1. After installation, go to Apps → Ascona AI in Jira’s top nav.
2. Click Sign in with Ascona AI (or similar button shown in the demo).
3. Authenticate with your Ascona AI account in the browser window that opens.
4. Approve the connection between Ascona AI and your Jira site.

4.2. Choose default project & settings

In the Ascona AI settings panel inside Jira, configure your defaults:

• Default Jira project: where new Epics & Stories will be created.
• Issue types: which types Ascona can create (Epic, Story, Task, Sub-task).
• Fields to auto-populate: description, acceptance criteria, labels, components, story points, etc.

Tip: Mirror your existing team conventions (labels, components, story point scale) so AI-generated work fits your current board with minimal editing.

5. Going from idea to structured backlog

The main flow shown in the demo is taking a plain-language idea and turning it into a full plan. Here’s how to reproduce that:

5.1. Open the Ascona AI panel

1. Navigate to the Jira project where you want to work.
2. Open the Ascona AI side panel (from Apps → Ascona AI, a project sidebar item, or the button shown in the demo).
3. You’ll see a prompt box where you can describe your project or feature.

5.2. Describe your project

Paste or type a high-level description, similar to what’s done in the video. For example:

5.3. Generate the initial plan

1. Click Generate plan (or the equivalent action button).
2. Wait a few seconds while Ascona AI:
   • Identifies major Epics (e.g., Workspace creation, Permissions, Activity feed).
   • Breaks each Epic into Stories and Tasks.
   • Drafts short descriptions for each item.
3. Review the proposed structure in the Ascona panel.

5.4. Refine before creating Jira issues

You can refine the draft plan directly from the Ascona AI panel:

• Regenerate: Ask Ascona to try again if the breakdown feels off.
• Adjust scope: Add context like “focus only on MVP” or “include analytics work.”
• Change detail level: Ask for “more granular tasks” or “fewer, higher-level tickets.”

Best practice: Iterate in the panel until the hierarchy looks right. It’s much easier than reorganizing dozens of tickets after they exist in Jira.

6. Creating Jira Epics, Stories, and Tasks

6.1. Pushing the plan into Jira

1. Once you’re happy with the generated plan, click Create issues in Jira (or the equivalent button).
2. Confirm:
   • The target Jira project.
   • Which levels to create (Epics only, Epics + Stories, or full hierarchy).
   • Whether to generate acceptance criteria for Stories.
3. Ascona AI will create all selected issues in bulk and show links to them.

6.2. How issues are structured

By default, the Jira issues created by Ascona AI usually follow this pattern:

• Epics: One per major workstream or feature area.
• Stories: User-facing slices or vertical work items within each Epic.
• Tasks / Sub-tasks: Technical implementation steps, testing, DevOps, etc.

6.3. Fields Ascona typically fills

• Summary: Clear, action-oriented ticket title.
• Description: Context, goals, and sometimes technical notes.
• Acceptance criteria: Bullet points describing “done.”
• Labels / Components: Optional, to help with reporting and filters.
• Story points / estimates: If enabled, AI can suggest a starting estimate.

Reminder: Treat AI estimates as suggestions. Your team should still do a quick groom to align points with your historical velocity.

7. Using Ascona AI on individual Jira issues

Beyond initial planning, you can use Ascona AI on specific Jira issues to improve clarity and quality. In the demo, this is shown via an Ascona AI panel or button on the issue view.

7.1. Expanding a rough ticket

1. Open any issue in Jira where the description is short or unclear.
2. Click the Ascona AI button or open the issue’s Ascona panel.
3. Choose an action such as:
   • Improve description – rewrite for clarity and completeness.
   • Generate acceptance criteria – based on summary & description.
   • Break down into subtasks – create a small checklist or sub-tasks.
4. Preview the proposed content, then click Apply to insert it into the issue.

7.2. Breaking a story into subtasks

1. Open a Story that feels too “chunky” for one person.
2. Use Ascona AI’s “Break down” option from the issue panel.
3. Optionally specify constraints, e.g. “3–6 subtasks” or “separate backend & frontend work.”
4. Review the suggested subtasks and confirm creation.

7.3. Clarifying technical scope

You can also give Ascona AI extra context right on the issue, such as:

• “We use React on the frontend and Node.js on the backend.”
• “We want this to be mobile-responsive and accessible.”

Ascona will then rewrite the description or acceptance criteria to reflect that stack and constraints, similar to what’s shown in the demo.

8. Using Ascona AI for sprint planning

Once your backlog is created, Ascona AI can help you prepare for sprint planning sessions.

8.1. Cleaning up the backlog

• Identify tickets missing descriptions, acceptance criteria, or estimates.
• Use Ascona AI on these tickets to fill in missing fields in bulk.
• Tag or label “AI-generated” items if you want to easily review them later.

8.2. Prioritization hints

Depending on your setup, Ascona AI can suggest rough ordering based on impact, dependencies, and user-value. Use it as a starting point, then have the product owner adjust priority.

8.3. Quick “what’s in this sprint?” summaries

For a selected set of issues (e.g., everything in the upcoming sprint), you can use Ascona AI to generate:

• A short summary paragraph of what the sprint is about.
• Key risks or unknowns spotted in the ticket descriptions.
• A checklist of cross-team dependencies.

Nice side effect: These AI summaries are great to paste into your sprint kickoff document, Confluence page, or stakeholder email.

9. Best practices for using Ascona AI in Jira

9.1. Write good prompts

• Describe target users, goals, constraints, and non-goals.
• Mention your tech stack if it affects implementation tasks.
• Call out what you do not want (e.g., “don’t include marketing tasks”).

9.2. Keep humans in the loop

• Review all generated tickets in grooming – especially estimates and acceptance criteria.
• Adjust naming conventions to match your team’s language.
• Use AI as a drafting partner, not an auto-pilot.

9.3. Iterate on your templates

If the output isn’t quite right, tweak your default prompt or internal guidelines. Over a few iterations, you’ll get plans that look like they were written by a senior PM.

10. Troubleshooting & FAQs

10.1. I don’t see the Ascona AI app in Jira

• Check that it’s installed under Apps → Manage apps.
• Confirm you’re in a project type supported by the app (usually Software projects).
• Ask your Jira admin if app access is restricted to certain groups or projects.

10.2. Ascona AI can’t create issues

• Make sure the app has permission to access the target project.
• Check if required fields (e.g., Components, Fix Version) are blocking issue creation.
• Try generating a simpler plan first to see if a specific ticket is causing errors.

10.3. The breakdown is too granular / not granular enough

• Adjust your prompt (e.g., “give me only 4–6 stories, no subtasks yet”).
• Use the refine/regenerate options shown in the demo before creating issues.
• Once you find a pattern you like, save that prompt as your “house style.”

10.4. Who should own Ascona AI inside the team?

Typically, a product manager, tech lead, or delivery manager owns the initial setup. But everyone on the squad (devs, QA, designers) can benefit from using Ascona AI on individual issues for clarity.