backlog-grooming

name	backlog-grooming
description	Weekly backlog audit — field completeness, staleness tiers, and duplicate detection. Run during grooming to clean up the backlog.
allowed-tools	["mcp__atlassian__jira_search","mcp__atlassian__jira_get_issue","mcp__atlassian__jira_get_project_components","mcp__atlassian__jira_get_project_versions","mcp__atlassian__jira_batch_get_changelogs","mcp__atlassian__jira_get_issue_dates"]

Backlog Grooming: Weekly Backlog Audit

Audit the full backlog for field completeness, staleness, and potential duplicates. Read-only — surfaces issues, never modifies tickets. Uses mcp__atlassian__* MCP tools exclusively (not jira-cli).

Custom Field Reference

Field ID	Name	Notes
customfield_10028	Story Points	The story points field for this instance
customfield_10464	Activity Type	Select dropdown

Important: Pass the fields parameter on every issue query:

fields: "summary,description,issuetype,status,priority,labels,assignee,reporter,created,updated,components,fixVersions,customfield_10028,customfield_10464"

Behavior

1. Pull the Backlog

Search using mcp__atlassian__jira_search with JQL:

project = HYPERFLEET AND status != Done AND status != Closed AND (sprint not in openSprints() OR sprint is EMPTY) ORDER BY updated ASC

Paginate if needed (use startAt parameter).

2. Field Completeness Checks

For each backlog issue, check these 6 fields:

Field	Check
Description	Exists and has meaningful content (> 20 chars)
Priority	Set to something other than the default
Story points	Assigned (for Stories and Bugs)
Component	Set and valid (verify against mcp__atlassian__jira_get_project_components)
Acceptance criteria	Present in description for Stories
Issue type	Not "Task" when it should be a Story or Bug

Score each issue: X/6 fields complete.

3. Staleness Tiers

Classify issues by time since last meaningful update:

Tier	Criteria	Action
Stale	No updates in 90+ days	Consider closing or re-evaluating
Aging	No updates in 60-89 days	Needs attention next grooming
Cooling	No updates in 30-59 days	Monitor
Active	Updated within 30 days	No action needed

Use mcp__atlassian__jira_batch_get_changelogs to check for meaningful updates (status changes, comment additions), not just field edits.

4. Duplicate Detection

Group issues by similarity:

• Compare summaries for near-duplicates (same keywords, similar phrasing)
• Flag issues in the same component with overlapping descriptions
• Present potential duplicates as pairs with a confidence note

This is heuristic — flag likely duplicates, don't assert.

5. Backlog Health Score

Calculate overall metrics:

• Total open issues in backlog
• Average field completeness (X/6)
• Distribution across staleness tiers
• Number of potential duplicates
• Issues with 0 story points (for Stories/Bugs)
• Fix Version coverage: for each issue, check the fixVersions array in the response — an issue has a Fix Version when this array contains at least one entry. Count issues with a non-empty fixVersions array vs those with an empty or missing array. Report this as informational only

Output Format

## Backlog Grooming Report
**[X] open issues in backlog** | Avg completeness: [X]/6

### Health Summary
| Metric | Count | % |
|--------|-------|---|
| Fully groomed (6/6) | X | Y% |
| Needs work (3-5/6) | X | Y% |
| Incomplete (0-2/6) | X | Y% |
| Has Fix Version | X | Y% |

### Staleness
| Tier | Count | Oldest |
|------|-------|--------|
| Stale (90+ days) | X | [PROJ-123] (X days) |
| Aging (60-89 days) | X | [PROJ-456] (X days) |
| Cooling (30-59 days) | X | |
| Active (< 30 days) | X | |

### Top Issues to Fix
1. [PROJ-123] — stale (120 days), missing description and points
2. [PROJ-456] — missing acceptance criteria, no component
3. [PROJ-789] — possible duplicate of [PROJ-012]
...

### Potential Duplicates
| Issue A | Issue B | Why |
|---------|---------|-----|
| [PROJ-123] | [PROJ-456] | Both reference "adapter timeout handling" |

### Incomplete Issues (0-2/6 fields)
| Ticket | Type | Summary | Missing |
|--------|------|---------|---------|
| [PROJ-123] | Story | ... | description, points, AC |

Integration

• hygiene: Complementary — hygiene validates specific tickets or sprint scope, this audits the full backlog
• ticket-triage: Use to walk through individual items that need grooming
• sprint-planning: Complementary — sprint-planning surfaces planning-relevant items, this gives the full backlog health picture

Notes

• Apply the ghostwriter skill for tone
• Cap "Top Issues to Fix" at 10 — prioritize by staleness + incompleteness
• Duplicate detection is best-effort. Use cautious language ("possible duplicate", "similar to")
• Do NOT use jira-cli or Bash for JIRA queries — use the mcp__atlassian__ MCP tools only