AI Prompt for Writing Clear Jira Bug Reports
Developers bounce bugs back when steps are vague or the expected result is missing. This prompt turns your raw notes into a Jira-ready report — and tells the AI to ask for missing facts instead of inventing them or guessing the root cause.
The prompt
You are a QA engineer writing a Jira bug report from rough notes. Rough notes: [PASTE YOUR NOTES] Environment (if known): [app version, build, browser/device, OS, environment URL placeholder] Evidence available: [screenshots, logs, HAR, video] Write the report using Jira wiki markup with these fields: - Summary: under 80 characters, "[Area] Problem when Condition" - Environment - Preconditions - Steps to Reproduce (numbered, one action per step) - Expected Result - Actual Result - Reproducibility (e.g. 3/3, intermittent 1/5) - Severity suggestion with one-line justification - Attachments to add Rules: - Use only facts from my notes. If something needed is missing, list it under "Missing information" instead of guessing. - Do not speculate on root cause. - Keep a neutral, factual tone.
Replace every [BRACKETED] placeholder with your own details.
Sample input: rough defect notes
Fictional example written for this guide.
checkout broke again?? on staging. added 2 items, applied code SAVE10, total went from 50 to 45 fine, then changed qty of 2nd item to 3 and discount disappeared but banner still says SAVE10 applied. tried twice, same. chrome 129 mac. didn't try safari. screenshot taken.
Sample AI output (illustrative)
Sample output only — this shows the shape of a good answer. It has not been run, executed, or verified against a real system.
*Summary:* [Checkout] Discount removed from total after changing item quantity *Environment:* Staging, Chrome 129, macOS (version not given) *Preconditions:* Promo code SAVE10 is active *Steps to Reproduce:* # Add two items to the cart (cart total = 50.00) # Apply promo code SAVE10 # Observe total updates to 45.00 # Change quantity of the second item to 3 *Expected Result:* Total is recalculated with the SAVE10 discount still applied. *Actual Result:* Discount is removed from the total, while the banner still shows "SAVE10 applied". *Reproducibility:* 2/2 attempts *Severity suggestion:* Major — customers may be charged more than the price shown as discounted. *Attachments:* screenshot (provided) *Missing information:* - Build / version number - Product names or SKUs used - Expected total after the quantity change - Behaviour on other browsers (Safari not tried)
Reproducibility checks before you file
- Reproduce at least twice from a clean state (new session or cleared cart).
- Note the exact ratio, e.g. 3/3 or 1/5 — “sometimes” is not actionable.
- Try one other browser or device to tell environment-specific from general bugs.
- Record the build number and timestamp so logs can be matched.
- Search Jira for duplicates using the key symptom, not your own wording.
Severity vs. priority guidance
Severity is the impact on the product; priority is how soon to fix. Teams define these differently — align with yours.
| Severity | Typical meaning | Example |
|---|---|---|
| Critical / Blocker | Core flow unusable, data loss, security issue, no workaround | Payment fails for all users |
| Major | Important feature wrong, workaround is painful | Discount silently dropped at checkout |
| Minor | Feature works with small inconvenience | Sort order resets after refresh |
| Trivial | Cosmetic, no functional impact | Misaligned icon in footer |
QA review before you use the output
- • Read the steps and follow them yourself — if you cannot reproduce from them, neither can a developer.
- • Remove AI-invented details (versions, totals, user names) that were not in your notes.
- • Redact customer data, tokens, and internal URLs from notes and logs before using a public AI tool.
- • AI output can be wrong or incomplete. A human tester owns the final result.
Related prompts
Need a different QA prompt?
Browse 50+ copy-ready prompts or build your own in seconds.