Sprint History Analyzer
Find trends across 4+ sprints from your export: say/do, velocity stability, scope creep, carryover, retro follow-through, forecasts.
Before & after
Same prompt, same agent — the only difference is the skill.
I can help you think through your sprint forecast. You'd want to look at your team's velocity over the last few sprints, estimate the remaining work, and see if there's enough time to complete it. There are different ways to approach this—some teams use historical average velocity, others adjust for known factors. I'd need to see your sprint data to give you a better forecast. Do you have your backlog estimates and the team's recent velocity numbers?
VELOCITY ANALYSIS: Last 8 Sprints Velocity history: Sprint 1: 34 points Sprint 2: 41 points Sprint 3: 38 points Sprint 4: 45 points Sprint 5: 39 points Sprint 6: 42 points Sprint 7: 43 points Sprint 8 (current): 28 points completed, 3 days left Average velocity: 40.5 points (8-sprint mean) Recent trend: Stable (no downward drift) MONTE CARLO FORECAST: Current Sprint Remaining work: 32 points Days remaining: 3 days Burndown trend: -7 points/day (recent pace) Confidence intervals: 50th percentile (median): 42 points by sprint end (2-point buffer) 85th percentile: 50 points (achievable with overtime) 15th percentile: 35 points (risk case) VERDICT: LIKELY TO SHIP (55% confidence at current pace) - Recommed: Move 2 low-priority items to next sprint as buffer - Critical path: Login feature unblocks 3 subtasks; prioritize review - Risk: One developer out tomorrow reduces velocity 12%; reassign 4 points RETROSPECTIVE THEMES (last 3 sprints) - Blocker: Code review turnaround 1.5 days average - Win: Spikes down 20% (better estimation) - Drag: Flaky tests fail 1 in 5 CI runs (adds rework) ACTION ITEMS 1. Pre-emptive code review on payment feature (earliest blocker) 2. Double-check QA test cases for Login; don't let scope creep 3. Temporarily reduce CI flakiness before sprint end
About this skill
One sprint tells you little; six tell you whether the team is improving, overcommitting, or drowning in mid-sprint requests. This skill reads your sprint export, cleans the known traps, and shows the trends with the math.
What it does
- Fixes export traps first: carried-over issues in several sprints, re-estimates, sub-task points, unestimated work
- Computes say/do ratio, velocity variation, scope added after start, carryover, and cycle-time percentiles
- Names at most three patterns, each with the sprints and numbers that show it
- Checks retro follow-through and groups recurring themes by root cause
- Forecasts 'when will this backlog be done?' as a 50%-85% range with Monte Carlo or a stated rough bracket
What you get
A multi-sprint scorecard, data notes, top patterns, a range-based forecast, and up to three recommendations with a two-sprint check for each.
Best for
Scrum masters, agile coaches, and engineering managers with a Jira, Linear, or CSV export of recent sprints.
SKILL.md preview
---
name: scrum-master
description: Analyze several past sprints from a Jira, Linear, or CSV export to find the trends one sprint hides - say/do ratio, velocity stability, scope added after start, carryover, blocker and cycle-time drift, and retro actions that never got done - then produce a multi-sprint team health scorecard and a probabilistic "when will this backlog be done?" forecast with ranges. Use when the user says "analyze our last 6 sprints", "are we getting better or worse?", "why is our velocity all over the place?", "when will we finish this epic?", "score our team health from the data", "same retro items keep coming back", or uploads a sprint export.
---
… (sign up to view the full skill)More business & ops skills
View all Business & Ops skills →Confluence and Jira Template Builder
Build one Confluence or Jira template people actually fill in: real-example structure, placeholders, roll-up reports, required fields.
Confluence Space Architect and Cleanup
Make a Confluence space findable: shallow page tree, label taxonomy, navigation macros, and a CQL audit of stale and orphaned pages.
Delivery Loop: Goal to Verified Close
Drive one deliverable from agreed goal and acceptance criteria to verified evidence and a real closeout, with nothing left orphaned.
Jira JQL, Board and Automation Expert
Correct JQL on the first try, plus boards, dashboards, workflows, and automation rules that avoid Jira's silent traps.