| name | sprint-report |
| description | Analyses Jira sprint data and produces a concise PM-ready report. Use whenever someone shares a sprint report, burndown chart, velocity chart, board screenshot, or Jira ticket data and wants to understand sprint health, delivery confidence, risks, blockers, or team performance. Trigger for phrases like "analyse this sprint", "what does the burndown say", "how is the sprint looking", "write up the sprint report", or "what should I tell the team today". |
| version | 1.1.0 |
| argument-hint | <sprint data, burndown, or Jira board link> |
| allowed-tools | Read, mcp__claude_ai_Atlassian_Rovo__searchJiraIssuesUsingJql, mcp__claude_ai_Atlassian_Rovo__getJiraIssue, mcp__claude_ai_Atlassian_Rovo__createConfluencePage, mcp__claude_ai_Atlassian_Rovo__getConfluenceSpaces |
Input
$ARGUMENTS
If no input is provided above, ask: "Please paste your Jira sprint data - burndown, velocity, ticket statuses, or a screenshot description. Which day of the sprint are we on?"
Minimum viable inputs: committed vs done, plus the sprint day or days remaining. If a paste is missing either, ask once, then proceed with whatever the user provides. If days remaining are unknown, do not emit a confidence percentage - write "Sprint Confidence: not assessable without days remaining" instead. If no velocity history is given, state that the forecast is based on this sprint only.
If Jira is connected (pull path): ask for the project key and the sprint name or ID. Validate the project key against [A-Za-z0-9_]+ before building JQL. Pull status, assignee, story points, and flagged state for the sprint's issues. Caveat: a query of current issues cannot reliably reconstruct the committed baseline or mid-sprint scope adds, so ask the user for the commitment figure and the sprint day rather than inferring them from the current backlog.
If Jira is not connected (MCP tool unavailable): do not fabricate ticket data. Tell the user Jira isn't connected and ask them to paste the board state (ticket statuses, assignees, blockers). The report format is identical - only the data source changes. See Connection Failsafe in .claude/CLAUDE.md.
Prior report check: when the active client/project is known, look for an earlier dated report on the same sprint in clients/CLIENT/PROJECT/sprint-artefacts/. If one exists, read it and open the Executive Summary with the movement since then - confidence shift, blockers still open or unmoved, new risks. Never search across clients. If no prior report exists or the project is ambiguous, proceed as normal.
Assess: burndown trend, completed vs committed, sprint goal attainment, blocked/aging tickets, scope changes, QA bottlenecks, overloaded owners, and velocity trend.
Sprint mode: determine whether the sprint is in flight or closed - from the input ("Day 10 of 10", "sprint ended", past-tense data) or by asking once. For a closed sprint, keep the format below with two swaps: Forecast becomes actuals (completed vs committed, velocity achieved), and Actions Today plus Questions for Standup are replaced by a single Carry-over and Next Sprint Implications section covering which items carry over, where they land, and what that means for the next commitment. Everything else is unchanged.
MANDATORY OUTPUT FORMAT
Executive Summary
3 short sentences maximum.
Delivery Status
- Status: Green / Amber / Red
- Sprint Confidence: XX%
- Forecast: X of Y items likely complete
- Velocity: committed X vs ~Y trailing average - on trend / over-committed / under-committed (only when velocity history is provided)
- Sprint goal: on track / at risk / missed - goal in a few words (if no goal was provided, write "not stated in input" - never invent one)
- Risk: Low / Medium / High
Top 3 PM Priorities
Main Risks
Actions Today
Questions for Standup
Leadership Update
2 short professional sentences.
FINAL ACTION (MANDATORY)
After generating the report, ask the user where they want it saved:
"Where would you like me to save this? I can save it locally or push it directly to any platform you have connected via MCP."
Local
- Local file - saved to
clients/CLIENT/sprint-artefacts/YYYY-MM-DD-sprint-N-report.md
Connected platforms (via MCP)
2. Confluence - published as a new page (I'll ask for your domain, space, and parent page)
3. Google Drive - saved as a new Doc (I'll ask for the folder)
4. Notion - created as a new page (I'll ask for your workspace and parent page)
No save
5. Clipboard only - leave it here for you to copy manually
Do not save or publish anything until the user confirms the destination.
If Confluence is chosen:
- Ask for the Confluence cloud domain, space key, and parent page title or ID
- Resolve the space ID using
getConfluenceSpaces if needed
- Use
createConfluencePage with:
cloudId: as provided by the userspaceId: resolved from the space keyparentId: as provided by the usertitle: {Project} Sprint Report - {Sprint Name}, Day {N} ({YYYY-MM-DD}) - for a closed sprint use Close-out in place of Day {N}contentFormat: markdownbody: the full report markdown
- Return the live Confluence page URL to the user
