Weekly Plan

Weekly Planning Coordinator

Every Friday or sprint transition, compile next week's plan by auditing Jira workload, reviewing team member capacity, and allocating high-priority items.

Iron Law: FOCUS OVER VOLUME. Each person gets up to 10 items max. The plan starts with a high-level summary (features, bugs), then drills into per-person details.

Step 0: Bootstrap

Read silently:

1. .env to load:
   • JIRA_PROJECT_KEY (referred to as PROJECT_KEY below)
   • CONFLUENCE_SPACE_KEY
   • CONFLUENCE_PARENT_PAGE_ID
2. team.json — full roster with domains, aliases, Jira accountIds.
3. Parse $ARGUMENTS:
   • --person name — plan for one specific team member only.
   • --no-publish — skip Confluence.

Constants:

• JIRA_CLI: python3 tools/jira-api.py

Build the person list from team.json. For each person, note their domains array for smart assignment.

Step 1: Gather Data

Run these Jira queries in parallel using python3 tools/jira-api.py search. Always pass --max-results 100 to get enough data for team-wide planning (default of 5 is insufficient).

Request fields that include timestamps for age-based analysis: add --fields summary,status,assignee,priority,labels,created,updated,resolved,issuetype,parent to each query.

# In Progress (carry-overs)
python3 tools/jira-api.py search --max-results 100 --fields summary,status,assignee,priority,labels,created,updated,issuetype,parent \
  "project = ${PROJECT_KEY} AND status = 'In Progress' ORDER BY priority DESC"

# To Do (candidates for next week)
python3 tools/jira-api.py search --max-results 100 --fields summary,status,assignee,priority,labels,created,updated,issuetype,parent \
  "project = ${PROJECT_KEY} AND status = 'To Do' AND priority in (Highest, High, Medium) ORDER BY priority DESC, created DESC"

# Recently completed (context for what shipped)
python3 tools/jira-api.py search --max-results 50 --fields summary,status,assignee,priority,resolved,issuetype,parent \
  "project = ${PROJECT_KEY} AND status = Done AND resolved >= -7d ORDER BY resolved DESC"

# Blocked items
python3 tools/jira-api.py search --max-results 50 --fields summary,status,assignee,priority,labels,created,updated,issuetype,parent \
  "project = ${PROJECT_KEY} AND status = 'In Progress' AND labels in (blocked, on-hold)"

# Unassigned items
python3 tools/jira-api.py search --max-results 50 --fields summary,status,assignee,priority,labels,created,issuetype,parent \
  "project = ${PROJECT_KEY} AND status in ('To Do', 'In Progress') AND assignee is EMPTY ORDER BY priority DESC"

Step 2: Analyze

High-Level Summary First

Before per-person plans, produce a team-wide summary:

1. Group issues by epic/parent — use the parent field to cluster related Jira issues under their epic.
2. Classify by type: features (Story/Task under epics), bugs (Bug type), tech debt, ops.
3. Summarize: "Next week: 3 new features in development, 2 bugs to fix, 1 tech debt item".
4. List features by name with their epics and constituent Jira issues (one feature/epic can have 3-4 child issues).

Per-Person Analysis

For each team member:

1. Carry-overs — list their In Progress items (these continue next week).
2. Capacity check — if carry-overs >= 10, flag as overloaded; do not add new items.
3. New priorities — from To Do backlog, match items to person's domains (from team.json). Suggest enough to fill up to ~10 total items.
4. Blocked items — flag any of their items that are blocked, with reason if available from labels/comments. Use the updated timestamp to compute age.

Flags to Surface

• Team members with 0 items (idle — need assignment).
• Team members with >10 items (overloaded — need re-prioritization).
• Blocked items >3 days old (compute from updated field — stale blocks).
• Unassigned items with High/Highest priority (need owners).
• Items carried over >2 weeks (compute from created/updated — chronic carry-over).

Step 3: Generate Plan

Console Summary (under 50 lines)

## Weekly Plan — {next week date range}

### Summary
- 3 features in development, 2 bugs to fix, 1 tech debt
- Features: Feature A (EPIC-${PROJECT_KEY}-XXX), Feature B (EPIC-${PROJECT_KEY}-YYY)
- Bugs: Crash fix (${PROJECT_KEY}-AAA)

### Feature: Feature A (Epic ${PROJECT_KEY}-XXX)
| Key      | Summary                    | Assignee | Status      |
|----------|----------------------------|----------|-------------|
| ${PROJECT_KEY}-XX01 | Backend handler            | Kavish   | In Progress |
| ${PROJECT_KEY}-XX02 | Android client             | Umesh    | To Do       |

### {Person Name} ({alias}) — {N items}
| # | Key      | Summary                          | Status      | Priority |
|---|----------|----------------------------------|-------------|----------|
| 1 | ${PROJECT_KEY}-XXXX | Carry-over: fix bug              | In Progress | High     |
| 2 | ${PROJECT_KEY}-YYYY | New: implement feature           | To Do       | Medium   |

### Flags
⚠️ Umesh has 12 items — needs re-prioritization
⚠️ ${PROJECT_KEY}-12345 blocked for 5 days — unblock before Monday
⚠️ 3 unassigned High-priority items need owners

Confluence HTML

Structured plan with:

• High-level summary at top: feature count, bug count, feature names with epics.
• Feature/epic sections: grouped Jira issues under each epic.
• Per-person sections: carry-over vs new items clearly marked, links to each Jira issue.
• Flags and recommendations at bottom.
• Week-over-week comparison: items planned last week vs completed.

Step 4: Deliver

1. Publish to Confluence (unless --no-publish):
   • Space: CONFLUENCE_SPACE_KEY, Parent: CONFLUENCE_PARENT_PAGE_ID.
   • Title: Weekly Plan — {date range}.
   • Format: HTML storage format.
2. Unassigned items — suggest assignments based on team.json domains.
3. Blocked items — suggest creating follow-up Jira comments to unblock.

Final: Report to User

Concise summary:

• High-level: "{N} features, {M} bugs, {P} tech debt items for next week"
• Feature names and their epics
• Number of carry-overs vs new items
• Any flags (overloaded, blocked, unassigned)
• Link to Confluence plan (omit if --no-publish)
• "Review and adjust assignments before Monday"