plan-sprint

Planning a Sprint

Select tickets, define a goal, and commit to a sprint scope.

Quick Reference

• Input: Prioritized backlog, team capacity, sprint duration.
• Output: Sprint plan with selected tickets, sprint goal, capacity allocation.
• Rule: Don't overcommit — plan for ≤80% of theoretical capacity.

HARD-GATE

DO NOT commit more than the team's historical velocity.
DO NOT select tickets with unresolved dependencies or unclear acceptance criteria.
DO set a single, clear sprint goal that every selected ticket supports.

Core Process

1. Gather inputs — prioritized backlog (from prioritize-backlog), team capacity, sprint dates, historical velocity.
2. Define sprint goal — every selected ticket must support this goal.
3. Select tickets — start from the top of the backlog, stop when total estimate reaches ≤80% of capacity (see Capacity Heuristics below).
4. Validate — check no ticket has unresolved dependencies, unclear acceptance criteria, or missing estimates. Use the Common Anti-Patterns table as a checklist.
5. Flag — tickets that almost made the cut (next sprint), dependency risks, stretch goals.
6. Output — sprint plan using the format in Output Style.

Capacity Heuristics

Theoretical Capacity

• Per person per day: 6 productive hours.
• Sprint capacity: team_size × sprint_days × 6h × utilisation_factor
• Utilisation factor: 0.8 for stable teams; 0.6–0.7 for new or high-uncertainty teams.

Velocity

• Rolling average: Mean completed points across last 3–5 sprints; exclude disrupted sprints.
• No history: Use theoretical_capacity × 0.6 as a conservative baseline.
• Declining trend: Use the most recent sprint's figure rather than the average.

Common Anti-Patterns

Anti-Pattern	Mitigation
Planning to 100% capacity	Cap at 80%
Including unestimated tickets	Estimate before committing
Multiple sprint goals	One goal per sprint
Ignoring dependencies	Resolve or defer dependents
Carry-over without re-estimation	Re-estimate carry-over tickets

Output Style

1. Sprint header — name, dates, duration, team capacity.
2. Sprint goal — single sentence.
3. Selected tickets — | Rank | Ticket | Estimate | Confidence | Notes |
4. Capacity allocation — total committed, buffer (%), stretch goals.
5. Deferred — top 3 items that didn't fit, for next sprint visibility.
6. English only unless user requests otherwise.

Worked example: assets/example-sprint-plan.md.

Integration

Skill	When to chain
prioritize-backlog	Use ordered backlog as input to ticket selection
estimate-tasks	Reference estimates for capacity-based selection
identify-risks	Scan sprint plan for dependency and capacity risks