Every prompt in this library follows a consistent structure. Understanding this structure helps you customise prompts for your own context.
ROLE → Who the AI should be
CONTEXT → What information to provide
TASK → What to do with that information
FORMAT → How to structure the output
TONE → How to communicate it
Sets the AI’s perspective and expertise level.
“You are an experienced Scrum Master facilitating Sprint Planning…”
Why it matters: AI responses are significantly better when given a specific expert lens. “Scrum Master with SAFe experience” produces different output than “helpful assistant.”
Customise it: If you’re a Release Train Engineer, adjust the role. If you’re coaching a junior SM, add "...explaining concepts clearly for someone new to the role".
The information the AI needs to produce relevant output. Always uses [brackets] for your team-specific inputs.
“Team: [team name], Sprint: [number], Completed stories: [list]”
Why it matters: Without real context, you get generic output. With specific context, you get output you can actually use.
Quality rule: The more specific your context, the less editing you’ll need to do on the output.
A clear, specific instruction for what to produce.
“Draft 2–3 candidate Sprint Goals that are outcome-focused, achievable within the sprint, and connect to the Product Goal.”
Why it matters: Vague tasks produce vague output. Specific tasks with clear requirements produce usable output.
Customise it: Add or remove requirements based on your team’s needs.
Specifies exactly how the output should be structured.
“For each candidate: Sprint Goal (one sentence), Why this goal (2–3 sentences), Key stories that deliver this goal”
Why it matters: Formatting instructions prevent AI from producing walls of text. They make output immediately usable in Confluence, Miro, or Teams.
Common formats used in this library:
Sets the communication register for the output.
“Professional, transparent, outcome-focused. Suitable for a Teams post.”
Why it matters: Tone changes everything. A retro summary for the team reads differently from a PI readout for Business Owners.
Tone options used in this library:
Conversational, team-focused — retro facilitationProfessional, stakeholder-friendly — sprint review outputsFactual, actionable — risk summaries, ART Sync notesClear, confident — planning summaries, PI objectivesUse this template to build new prompts:
You are a [ROLE with relevant expertise].
CONTEXT:
- [Key piece of context 1]: [bracket for input]
- [Key piece of context 2]: [bracket for input]
- [Add as many as needed]
TASK:
[Clear instruction for what to produce. Include requirements as a numbered or bullet list.]
FORMAT:
[Specify exactly how you want the output structured.]
TONE: [Describe the communication register and intended audience.]
Have a prompt that works well for your team? Contribute it — the structure above is the standard to follow.