Feature Review

Review currently implemented features using evidence-based prioritization and suggest new features. Accepted suggestions can be uploaded to GitHub as issues.

Usage

# Full review: inventory, score, suggest
/feature-review

# Only inventory current features
/feature-review --inventory

# Generate new feature suggestions
/feature-review --suggest

# Create GitHub issues for accepted suggestions
/feature-review --suggest --create-issues

# Validate configuration file
/feature-review --validate-config

# Override configuration values
/feature-review --threshold.high_priority=3.0

What It Does

1. Discovers implemented features from codebase
2. Classifies features as proactive/reactive and static/dynamic
3. Scores features using hybrid RICE+WSJF framework
4. Analyzes tradeoffs across quality dimensions
5. Suggests new features based on gaps and opportunities
6. Creates GitHub issues for accepted suggestions

Workflow

Phase 1: Feature Inventory

Scan the codebase to discover implemented features:

• Commands, skills, agents, hooks
• Public APIs and exports
• Documented features in README/CHANGELOG

Phase 2: Classification

Classify each feature along two axes:

Type Axis:

• Proactive: Anticipates user needs (suggestions, prefetching)
• Reactive: Responds to explicit input (commands, forms)

Data Axis:

• Static: Updated incrementally (configs, schemas)
• Dynamic: Updated continuously (sessions, real-time data)

Phase 3: Scoring

Apply hybrid prioritization scoring:

Score = (Value Score / Cost Score) * Confidence

Value Factors: Reach, Impact, Business Value, Time Criticality Cost Factors: Effort, Risk, Complexity

Thresholds:

• > 2.5: High priority (implement soon)
• 1.5 - 2.5: Medium priority (roadmap)
• < 1.5: Low priority (backlog)

Phase 4: Tradeoff Analysis

Evaluate across nine quality dimensions:

• Quality, Latency, Token Usage
• Resource Usage, Redundancy, Readability
• Scalability, Integration, API Surface

Phase 5: Suggestions

Based on analysis, suggest:

• New features addressing gaps
• Improvements to low-scoring features
• Deprecation of obsolete features

Phase 6: GitHub Integration

For accepted suggestions:

1. Generate issue from template
2. Apply priority and classification labels
3. Prompt for user confirmation
4. Create issue via GitHub CLI

Arguments

Argument	Description
--inventory	Only run feature discovery, skip scoring
--suggest	Include new feature suggestions
--create-issues	Create GitHub issues (requires --suggest)
--validate-config	Validate configuration file
--dry-run	Show what would be created without creating

Configuration

Create .feature-review.yaml in project root to customize:

• Scoring weights and thresholds
• Classification patterns
• Tradeoff dimension weights
• GitHub integration settings

See Skill(imbue:feature-review) for full configuration reference.

Examples

Full Review

/feature-review

# Output:
Feature Review Report
=====================

## Inventory (12 features found)

| Feature | Type | Data | Score | Priority |
|---------|------|------|-------|----------|
| skill-loader | Reactive | Static | 2.8 | High |
| auto-complete | Proactive | Dynamic | 2.3 | Medium |
| ...

## Tradeoff Analysis

skill-loader:
  Quality: 5/5 | Latency: 4/5 | Token Usage: 3/5
  Readability: 4/5 | Integration: 5/5 | API Surface: 5/5

## Suggestions

1. [High] Add caching for skill lookups (Score: 2.6)
2. [Medium] Improve error messages (Score: 1.9)

Create Issues

/feature-review --suggest --create-issues

# Output:
Suggestions ready for GitHub:

1. Add caching for skill lookups
   Score: 2.6 | Classification: Proactive/Static
   Labels: enhancement, priority/high, type/proactive

   Create issue? [y/n]: y

   Created: https://github.com/owner/repo/issues/123

Integration

This command uses:

• Skill(imbue:feature-review) - Core scoring and classification
• Skill(imbue:scope-guard) - Budget validation for suggestions
• Skill(imbue:review-core) - Evidence-based methodology

Exit Criteria

• Feature inventory complete
• All features classified and scored
• Tradeoff analysis documented
• Suggestions generated (if requested)
• GitHub issues created (if requested)