product-owner

What this skill does

# Product Owner Skill

A Claude-native product owner that provides daily standups, priority recommendations, and cross-project awareness without artificial sprint boundaries. Enhanced with AI-powered analysis, metrics tracking, and intelligent dependency detection.

## Contextd Integration (Optional)

If contextd MCP is available:
- `memory_record` for standup patterns, blockers, and prioritization decisions
- Cross-session priority tracking with decision accuracy measurement
- `memory_search` to find past standup context and recurring patterns
- `remediation_search` for known blockers and their resolutions

If contextd is NOT available:
- GitHub queries work normally
- No cross-session memory (stateless standups)
- No decision tracking or pattern learning

## Philosophy

- **Recommend, don't dictate**: Present priorities; user decides what to act on
- **Continuous flow**: No sprints or ceremonies - priorities adjust daily
- **Cross-project awareness**: Dependencies across repos get flagged and prioritized
- **Memory persistence**: Use contextd if available (optional)
- **Data-driven decisions**: Leverage DORA metrics and WSJF for objective prioritization
- **Learn from history**: Track decision accuracy to improve recommendations over time

## When to Use

| Trigger | Use Case |
|---------|----------|
| `/standup` | Daily standup report with AI-generated blockers |
| `/standup --metrics` | Include DORA metrics and velocity |
| `/standup --platform` | Cross-project view with dependency map |
| "what should I work on?" | WSJF-ranked priority recommendations |
| "project status" | Current state with risk scoring |
| "what's blocking?" | AI-powered blocker analysis |
| "show velocity" | Velocity tracking and forecasting |
| "show metrics" | DORA metrics dashboard |

## Data Sources

### GitHub (via MCP)

| Query | Purpose |
|-------|---------|
| `list_pull_requests` | Open PRs, review states, ages |
| `list_issues` | Priority-labeled issues |
| `list_commits` | Recent activity on main |
| `list_branches` | Detect stale branches |
| `get_commit` | Deployment tracking for DORA |
| `search_issues` | Cross-repo dependency detection |

### GitHub Projects v2 (via GraphQL)

| Query | Purpose |
|-------|---------|
| Project items | Sprint/iteration tracking |
| Custom fields | Story points, priority, status |
| Project views | Board and table layouts |

**GraphQL Example:**
```graphql
query {
  organization(login: "fyrsmithlabs") {
    projectV2(number: 1) {
      items(first: 100) {
        nodes {
          content { ... on Issue { title number } }
          fieldValues(first: 10) {
            nodes { ... on ProjectV2ItemFieldNumberValue { number } }
          }
        }
      }
    }
  }
}
```

### contextd (Cross-Session Memory)

| Query | Purpose |
|-------|---------|
| `checkpoint_list/resume` | Yesterday's state |
| `memory_search` | Recurring patterns, blockers |
| `remediation_search` | Known issues and fixes |
| `memory_record` | Store prioritization decisions |

---

## AI-Powered Analysis

### Risk Scoring

Calculate risk score (0-100) based on technical complexity and business impact:

```
Risk Score = (Technical Complexity * 0.4) + (Business Impact * 0.4) + (Time Sensitivity * 0.2)

Technical Complexity factors:
- Lines changed: >500 = HIGH, >100 = MEDIUM, else LOW
- Files touched: >10 = HIGH, >3 = MEDIUM, else LOW
- Cross-service changes: +20 points
- Database migrations: +25 points
- Security-sensitive areas: +30 points

Business Impact factors:
- Label priority:critical = 40, priority:high = 30, else 10
- Customer-facing: +20 points
- Revenue-impacting: +25 points
- Compliance-related: +30 points

Time Sensitivity factors:
- Days until deadline (if any)
- External dependencies timing
- Release train alignment
```

**Risk Categories:**
| Score | Category | Action |
|-------|----------|--------|
| 80-100 | CRITICAL | Immediate attention, may need escalation |
| 60-79 | HIGH | Prioritize this session |
| 40-59 | MEDIUM | Schedule within week |
| 0-39 | LOW | Normal queue |

### Semantic Dependency Detection

AI analyzes issue/PR content for implicit dependencies beyond explicit references:

**Patterns detected:**
- Shared code paths (same files modified)
- Common data models or schemas
- API contract dependencies
- Feature flag interdependencies
- Deployment order requirements

**Example detection:**
```
Issue #45: "Add user avatar upload"
Issue #52: "Implement image resizing service"
  -> DETECTED: #45 likely depends on #52 (image processing capability)
```

### Automated Blocker Identification

Scan issue comments and PR reviews for blocker signals:

**Signal patterns:**
```regex
# Explicit blockers
blocked by|waiting on|depends on|can't proceed

# Implicit blockers (in comments)
need.*first|requires.*before|after.*merged
stuck on|no progress|help needed

# Review blockers
changes requested.*\d+ days ago
approved but.*merge conflict
CI.*failing.*\d+ hours
```

**AI Comment Analysis:**
```
PR #42 Comments Analysis:
  - @alice (2 days ago): "This needs the auth refactor first"
  - @bob (1 day ago): "Still waiting on API spec"
  -> BLOCKERS DETECTED: Auth refactor, API spec finalization
```

---

## WSJF Prioritization

Weighted Shortest Job First for objective prioritization:

```
WSJF Score = Cost of Delay / Job Size

Cost of Delay = User Value + Time Criticality + Risk Reduction

Where:
- User Value: Business impact (1-10)
- Time Criticality: Urgency decay (1-10)
- Risk Reduction: Technical/business risk mitigated (1-10)
- Job Size: Estimated effort in story points or T-shirt sizes
```

### WSJF Calculation Example

```
Issue: Implement OAuth2 login
  User Value: 8 (high user demand)
  Time Criticality: 6 (competitor launching similar)
  Risk Reduction: 7 (security improvement)
  Job Size: 5 (medium complexity)
  WSJF = (8 + 6 + 7) / 5 = 4.2

Issue: Fix typo in docs
  User Value: 2 (minor inconvenience)
  Time Criticality: 1 (no deadline)
  Risk Reduction: 1 (no risk)
  Job Size: 1 (trivial)
  WSJF = (2 + 1 + 1) / 1 = 4.0

Result: OAuth2 login prioritized higher despite larger size
```

### Auto-WSJF from Labels

Map GitHub labels to WSJF components:

| Label | Component | Value |
|-------|-----------|-------|
| `priority:critical` | Time Criticality | 10 |
| `priority:high` | Time Criticality | 7 |
| `security` | Risk Reduction | 9 |
| `customer-facing` | User Value | 8 |
| `tech-debt` | Risk Reduction | 5 |
| `size:XL` | Job Size | 13 |
| `size:L` | Job Size | 8 |
| `size:M` | Job Size | 5 |
| `size:S` | Job Size | 3 |
| `size:XS` | Job Size | 1 |

---

## DORA Metrics Integration

Track and display key DevOps Research and Assessment metrics:

### Metrics Definitions

| Metric | Definition | Calculation |
|--------|------------|-------------|
| **Deployment Frequency** | How often code deploys to production | Deploys per day/week |
| **Lead Time for Changes** | Time from commit to production | PR open -> merge -> deploy |
| **Mean Time to Recovery** | Time to restore service after incident | Incident open -> resolved |
| **Change Failure Rate** | % of deployments causing failures | Failed deploys / total deploys |

### Data Collection

**Deployment Frequency:**
```
# Count merges to main/production in time period
mcp__MCP_DOCKER__list_commits(
  owner: "<org>",
  repo: "<repo>",
  sha: "main",
  since: "<period_start>"
)
```

**Lead Time for Changes:**
```
# For each merged PR:
Lead Time = PR merged_at - first_commit_at + deploy_delay
```

**Change Failure Rate:**
```
# Track labels/issues indicating failures
mcp__MCP_DOCKER__list_issues(
  labels: "incident,hotfix,rollback"
)
```

### Metrics Display Format

```
+-------------------------------------------------------------+
| DORA Metrics: <repo> (Last 30 days)                         |
+-------------------------------------------------------------+
| Deployment Frequency:  2.3/day  ########-- Elite            |
| Lead Time for Changes: 4.2 hrs  #######--- High             |
| Mean Tim