Building Spike Tickets

A spike is a time-boxed investigation ticket whose goal is to reduce uncertainty before committing to a solution. It answers a specific technical question — it does not deliver functionality. When the spike is done, you close it and create follow-up stories based on what you learned.

spike vs spike-output: use spike to create the ticket (the investigation plan). Use spike-output to document the findings once the investigation is complete.

Generate minimal Jira Spike tickets. Output in English, Jira Markdown, inside a code block.

When to Create a Spike

Create a spike when the team cannot estimate a story because something is unknown:

• Technology feasibility: "Can we integrate X with Y?"
• Architecture uncertainty: "Which of these two approaches should we take?"
• Performance unknowns: "Will this scale to N requests?"
• Third-party API exploration: "What does this API actually support?"
• Security or compliance questions: "How does the library handle auth?"

A spike is not needed when the team already knows how to do the work. If the answer is obvious, skip the spike and write the story directly.

Output Format

1. List 3 title suggestions outside the code block (Spike: [Action] [Subject] for [Context])
2. Output the ticket in a single code block using this template:

h2. Problem Statement
[2-3 sentences: what is unknown and why it matters.]

h2. Research Questions
* [Question 1]
* [Question 2]
* [Question 3]

h2. Activities
* [ ] [Task 1]
* [ ] [Task 2]
* [ ] [Task 3]

h2. Definition of Done
* Go/No-Go decision with rationale.
* Refined follow-up ticket ready for sprint planning (if a bug or action is needed).

Keep all sections brief. No filler text.

After the Spike

Once the investigation is done, use the spike-output skill to document findings and conclusions in a format ready to paste into the Jira ticket.