Generate a data-grounded sprint retrospective for the most recently closed sprint: What went well, What did not, Action items, and Discussion prompts. Every line is backed by a metric, ticket, PR, or Slack thread, framed around the process not individuals. Recurring themes across recent sprints are flagged.

The hard numbers behind the retro for the closed sprint: completion, mid-sprint scope change, carryover, reopens, average cycle time vs the prior sprint, velocity, PR review latency, reverts and hotfixes, and flagged Slack threads. This is the evidence base the brief draws from.

Open questions for the retro board, each grounded in a real anomaly from the closed sprint (a reopened ticket, mid-sprint scope, a slow review, a revert, a blocker). Questions to start the conversation, not verdicts that end it.

Close the loop on last retro's action items. Give the previous retro's action items and it attaches THIS sprint's relevant data to each, so the team can see what actually moved. Reviewing prior actions is the single proven lever on retro follow-through. It shows evidence only and lets you judge what counts as done.

Compare the closed sprint against the one before it (completion, scope, carryover, reopens, cycle time, velocity), and list the problem themes that recur across recent sprints. Use it to tell a one-off from a pattern.

Show which data sources are configured (Jira, plus optional GitHub and Slack enrichment), or that the server is running on a synthetic demo sprint. Does not hit the network; run the server with --check to verify live connections.