AI agent for scrum masters
Carry-Over and Scope Change Analysis Agent
Find the main cause of carry-over and test one fix at a time
What it does
The same stories carry over sprint after sprint and each retrospective names a new reason. This agent reads several sprints of history and classifies each carry-over by cause: the story was too big, it was blocked, scope was added mid-sprint, the team was short staffed, or the plan was too ambitious. It counts each cause across sprints to find the main pattern. It proposes one experiment, for example a story size limit or a cap on mid-sprint additions, with a measurement. It then checks the next sprint's numbers against the target and reports whether the experiment worked. The team approves the experiment. Edge case: causes overlap, such as a large story that was also blocked, so the agent assigns a main cause and a second one.
How it works
Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.
Read the steps as a list
- Sprint review finished
- Read the last 6 sprints and the carried-over stories
- Classify each carry-over by main and second cause
- Count causes across sprints and find the main pattern
- Propose one experiment with a target
- Team approves the experimentThe agent waits here for your OK.
- Read the next sprint's data
- Did the carry-over rate fall toward the target?If not: check whether the experiment was followed and propose an adjustment or a different one. Back to step 4.
- Record the result and the next pattern
- Carry-over analysis and experiment result
How it decides
It assigns each carried-over story a main cause from the history and proposes an experiment for the most frequent one.
- Treat a story as large if over 8 points or open over 5 days
- Count a story as blocked when its status was blocked for over 1 day
- Count scope added after day 2 as scope change
- Test one experiment at a time
Make it yours
Every agent is a starting point. You choose these settings for your own situation.
- Sprints analyzed (default 6)
- Cause categories
- Story size limit
- Experiment target
- Report frequency
What keeps you in control
It always asks you first
- Team approves each experiment
- Scrum master approves sharing the analysis
Hard limits
- Never blame individuals in the analysis
- Never change stories in the tracker
It stops when
- Done: carry-over is within target for two sprints
- Stop: data is too thin to find a pattern
Set it up
We guide you through the set-up, step by step
Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.
- One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
- The agent then walks you through connecting your own data, one source at a time
- A downloadable copy with the flow chart, the rules and the full guide
