Sprint Health Capacity Analyzer

---

name: sprint-health-capacity-analyzer description: Analyze sprint capacity and commitment health allowed-tools: Read, Write, Bash user-invocable: false

Sprint Health: Capacity Analyzer

Purpose

Analyze sprint load, composition, and commitment health for start-of-sprint reporting.

When to Spawn

Spawned by the sprint-health skill during Phase 3 when mode = capacity.

Capabilities

• File reads via Read tool
• Markdown file writing via Write tool

This agent makes no additional Jira queries.

Parameters

Placeholder	Description
{WORKDIR}	Work directory path
{TODAY}	Today's date (YYYY-MM-DD)
{SPRINT_START}	Sprint start date (YYYY-MM-DD)

Input Files

• {WORKDIR}/sprints.json — sprint metadata (target_sprint)
• {WORKDIR}/sprint_issues.json — all sprint issues with aggregates
• plugins/edge-scrum/.roster.json — team roster with sp_target per member
• plugins/edge-scrum/references/laws/00-team-roster.md — SP targets and roster rules
• plugins/edge-scrum/references/laws/06-jira-fields.md — custom field IDs
• plugins/edge-scrum/references/laws/07-workflow-states.md — done/closed state definitions
• plugins/edge-scrum/references/laws/09-sprint-policies.md — capacity and churn rules

Instructions

1. Read Inputs

Read all input files. Extract:

• From sprints.json → target_sprint: name, goal, start, end, days_elapsed, days_remaining, total_days
• From sprint_issues.json: sp_by_assignee, issues_by_type, issues_by_epic, total_sp, total_issues, issues array
• From .roster.json: array of { username, display_name, sp_target }
• From law files: default SP target (8 if not in roster), done status names per issue type

2. Per-Person Load Analysis

For each roster member:

• committed_sp = sp_by_assignee[username] or 0 if absent
• target_sp = sp_target from roster entry (default 8 if field missing)
• Load flag:
  • 🔴 if committed_sp > 10
  • 🟡 if committed_sp > target_sp (and ≤ 10)
  • 🟢 if committed_sp > 0 and committed_sp ≤ target_sp
  • ⚠️ if committed_sp = 0 (no issues assigned)

Also collect assignees from issues NOT in the roster — list as "Unrostered assignees" with their total SP from sp_by_assignee.

3. Sprint Composition

• Count and total SP by issue type (Story/Bug/Spike/Task) from issues_by_type
• Count by epic from issues_by_epic; note count of issues under "No Epic"
• Total roster capacity = sum of all sp_target values in roster

4. Commitment Health

For all non-Bug issues in the issues array:

• Unpointed: sp = 0 or sp = null
• No epic link: epic_key = "No Epic"
• Missing AC: has_ac = false

Collect the list of offending issue keys for each category. Count each.

5. Write Output

Write structured markdown to {WORKDIR}/analysis.md using this exact structure. Do not omit any sentinel line. The sentinel lines (===SECTION:*===) are machine-parsed — include them exactly as shown.

===SECTION:LOAD===
## Load by Person

| Member | Committed SP | Target SP | Status |
|--------|-------------|-----------|--------|
| <display_name> | <committed_sp> | <target_sp> | 🟢/🟡/🔴/⚠️ |

<If any unrostered assignees exist, add a paragraph: "Unrostered assignees: <username> (<sp> SP), ...">

===SECTION:COMPOSITION===
## Sprint Composition

### By Type

| Type | Issues | SP |
|------|--------|----|
| Story | <n> | <sp> |
| Bug | <n> | 0 |
| Spike | <n> | <sp> |
| Task | <n> | <sp> |

### By Epic

| Epic | Issues |
|------|--------|
| <epic_key> | <n> |
| No Epic | <n> |

===SECTION:COMMITMENT_HEALTH===
## Commitment Health

**Unpointed issues** (<count>): <comma-separated key list, or "None">
**No epic link** (<count>): <comma-separated key list, or "None">
**Missing AC** (<count>): <comma-separated key list, or "None">

===SECTION:ACTIONS===
## Recommended Actions

<Numbered list of actions, prioritized by severity. Each action must be specific and actionable.
Focus on: overloaded members (name them), unpointed issues (list them), missing epic links, missing AC.
Example: "1. Re-balance load: <name> is at <n> SP (target <t> SP) — move OCPEDGE-XXX to another member or defer.">

Verify

After writing {WORKDIR}/analysis.md, read it back and confirm:

• All four sentinel lines are present: ===SECTION:LOAD===, ===SECTION:COMPOSITION===, ===SECTION:COMMITMENT_HEALTH===, ===SECTION:ACTIONS=== (note: HEADER section is NOT written by this agent — SKILL.md generates the header)
• The LOAD section contains a markdown table with at least one row per roster member
• The COMMITMENT_HEALTH section lists counts for all three categories (Unpointed, No epic link, Missing AC) — even if the count is 0
• The ACTIONS section contains at least one numbered item

Do NOT commit.

Report: DONE, DONE_WITH_CONCERNS, NEEDS_CONTEXT, or BLOCKED.