Automated Sprint Planning Agent

The short version

The Sprint Planning agent runs every Monday at 9 AM and converts your prioritized opportunities into a proposed sprint plan in 30 minutes instead of 4 hours. It reads opportunities, tech debt backlog, team capacity, and historical velocity from Jira or Linear, then breaks each opportunity into user stories with story points, fits stories into the sprint respecting capacity and dependencies, and explains every trade-off. The output is ready to load into Jira, with a confidence score on the plan and a list of which stories are most likely to slip. Half a day reclaimed, every sprint.

Sprint planning takes half a day. You sit with your list of prioritized opportunities, break them into stories, estimate effort, account for team capacity and tech debt, and move pieces around until something resembles a plan.

By the end you're exhausted and you've probably forgotten half the context that made these opportunities matter.

The Sprint Planning agent does this on Monday mornings. It reads: prioritized opportunities, tech debt backlog, team capacity, and sprint constraints. It produces: a proposed sprint plan with stories, estimates, and explanations of why each item made the cut.

You spend 30 minutes reviewing instead of 4 hours planning.

How It Works

The agent takes structured inputs and applies logic in sequence:

Story generation: For each opportunity, the agent breaks it into user stories. Not generic stories, but specific: "As a [segment] user, I can [do X], so that [outcome]." Including acceptance criteria and technical tasks.

Effort estimation: Using historical velocity and similar stories, the agent estimates story points for each. It flags estimates that feel risky (too much guessing).

Sprint composition: The agent fits stories into the sprint, respecting: team capacity, committed tech debt time, and dependencies. It surfaces: "If we do opportunity A, we have to do tech debt B first. That's 13 points. We have 21 points capacity, so we can add one more opportunity."

Explanation: For each sprint plan, the agent explains: why these opportunities, why not the others, and what trade-offs were made.

Data Sources and Setup

Prerequisites: You'll need:

• Prioritized opportunities: From the Opportunity Prioritization agent
• Jira or Linear: For story backlog, historical velocity, tech debt items
• Team capacity data: Planned time off, support rotations, meetings
• Feature specs and tech context: For realistic effort estimation
• Previous sprints: Historical velocity and story estimation patterns

Schedule: Weekly Monday at 9 AM. Can also run ad-hoc when priorities change.

The Claude Prompt

You are planning our sprint based on prioritized opportunities.

Here are our top prioritized opportunities:
[OPPORTUNITIES: ranked, with impact and effort estimates]

Here's our tech debt backlog:
[TECH DEBT: items, severity, impact on velocity]

Here's our team capacity for the next sprint:
[CAPACITY: team size, planned time off, support load, committed meetings]

Here's our historical context:
[HISTORY: past 4 sprints velocity, similar story estimates, completion rates]

Here are feature specs and dependencies:
[SPECS: technical requirements, dependencies between features]

Please build a sprint plan that includes:

1. **Story Breakdown**
   - For each top 3-5 opportunities, break into 3-5 stories each
   - Format: "As a [role], I can [action], so that [outcome]"
   - Include acceptance criteria and technical tasks
   - Estimate story points using historical patterns

2. **Tech Debt Assessment**
   - Which tech debt items should we include?
   - Why? (blocks other work, affects velocity, or critical for stability?)
   - How much capacity should we allocate?

3. **Sprint Plan**
   - Show the proposed sprint with story points
   - Total capacity vs. committed points
   - Explain why these opportunities made the cut
   - Show the prioritized backlog (what didn't make it and why)

4. **Risk and Dependencies**
   - Are any stories dependent on each other? (order matters)
   - Are any stories dependent on external teams?
   - Which stories carry the most uncertainty in estimation?
   - What could make this sprint go wrong?

5. **Confidence Assessment**
   - How confident are we in this plan? (% chance we complete as planned)
   - What would improve confidence?
   - Which stories are most likely to slip?

Format as a sprint proposal. Be specific: I should be able to load this into Jira and start working.

What You Get

Instead of 4-hour planning sessions:

• Structured sprint plan: Ready to load into Jira, with stories and estimates
• Opportunity-driven: You're building what matters, not whatever the loudest person wanted
• Capacity-aware: The plan respects your team's actual capacity (including tech debt and off-time)
• Dependency mapping: You know what order to tackle things
• Risk visibility: You know which stories might slip and why

Real outcomes:

• Sprint planning takes 30 minutes instead of 4 hours
• Your team spends less time debating scope and more time building
• You actually complete your sprints because they're realistic

For the full agent fleet and scheduling details, see Your AI Agent Fleet.