Product Management

AI-native product management for startups. Transform Claude into an expert PM that processes signals, not just feature lists.

Features

• Competitive Research: Deep autonomous competitor analysis with multi-source research
• Gap Analysis: Systematic identification with WINNING filter prioritization
• PRD Generation: Generate PRDs and automatically create GitHub Issues
• Deduplication: Sync with GitHub Issues to avoid duplicate feature requests
• Staleness Alerts: SessionStart hook warns when PM data is outdated

Core Philosophy

WINNING = Pain × Timing × Execution Capability

Score: 40-60 FILE | 25-39 WAIT | 0-24 SKIP

Installation

# Add to Claude Code
claude plugin add /path/to/product-management

# Or copy to project
cp -r product-management/ /your/project/.claude-plugin/

Commands

Commands Reference

/pm:analyze - Product Discovery

Purpose: Build a comprehensive understanding of your current product

When to use: First command when starting PM workflow on a new project

What it does:

• Scans codebase structure (domains, APIs, features)
• Interviews you about positioning, target users, differentiators
• Creates .pm/product/inventory.md with feature inventory
• Identifies technical moats and constraints

Output: .pm/product/ directory with product documentation

/pm:landscape [name] - Competitor Research

Purpose: Research the competitive landscape or deep-dive on specific competitors

When to use: After /pm:analyze, before gap analysis

Usage:

/pm:landscape              # Market overview, identify top 5 competitors
/pm:landscape Expensify    # Deep-dive on specific competitor
/pm:landscape Dext         # Deep-dive on another competitor

What it does:

• Without argument: Creates market landscape overview
• With argument: Deep research on named competitor (pricing, features, positioning)

Output: .pm/competitors/_landscape.md or .pm/competitors/[name].md

/pm:gaps - Gap Analysis with WINNING Filter

Purpose: Identify feature gaps and score them objectively

When to use: After competitor research, to prioritize what to build

What it does:

1. Loads product inventory + competitor data
2. Identifies gaps (features competitors have that you don't)
3. Scores each gap with WINNING filter (Pain + Timing + Execution + Fit + Revenue + Moat)
4. Outputs decision: FILE (40-60), WAIT (25-39), SKIP (0-24)

Output: .pm/gaps/[date]-analysis.md with scored gaps and recommendations

/pm:file [id|all] - Batch File Gaps as Issues

Purpose: Convert gaps from analysis into GitHub Issues (lightweight)

When to use: After /pm:gaps when you want to quickly file multiple gaps

Usage:

/pm:file              # Interactive: review each gap, decide FILE/WAIT/SKIP
/pm:file GAP-003      # File specific gap by ID
/pm:file all          # Batch file all gaps scored 40+ (FILE threshold)

What it does:

• Creates GitHub Issues from gap analysis results
• Applies labels: pm:feature-request, winning:high/medium/low, priority:now/next/later
• Checks for duplicates against synced issues

Output: GitHub Issues with gap summary, WINNING score breakdown

Issue detail level: Brief (title, score, short description)

/pm:prd - Generate Full PRD + Issue

Purpose: Create comprehensive Product Requirements Document for a specific feature

When to use: When you need detailed specs before implementation

Usage:

/pm:prd Plaid bank transaction matching
/pm:prd Bulk invoice processing
/pm:prd Real-time collaboration

What it does:

1. Gathers context from gap analysis and competitor data (if available)
2. Generates comprehensive PRD with 10+ sections
3. Saves PRD to .pm/prds/[feature-slug].md
4. Creates GitHub Issue with full PRD as body

PRD sections:

• Problem Statement
• User Stories
• Competitive Analysis
• Requirements (Functional, Non-Functional)
• Acceptance Criteria (P0/P1/P2)
• Edge Cases
• Out of Scope
• Technical Considerations
• Success Metrics

Output: Full PRD file + detailed GitHub Issue

Issue detail level: Comprehensive (full PRD, ready for spec-kit handoff)