By abhitsian
Flywheel PM — 13 agents, 32 commands, 13 skills. Small pushes, unstoppable momentum. Your PM operating system that gets sharper with every product cycle.
Check a product decision against cognitive biases most likely to affect your judgment right now
Generate stakeholder-specific alignment narratives using proven buy-in techniques
Capture learnings from a product cycle and update your PM profile — the most important step
Set up and update your work context — company priorities, team capacity, customer signals, competitive landscape
Frame a decision explicitly with options, criteria, trade-offs, reversibility, and stakeholders before committing
Proactively detects cognitive biases in product decisions, opportunity evaluations, and specs. Use to surface biases before they distort decisions.
Reviews teardowns and simulations for blind spots, bias patterns, and simulation accuracy. Use when evaluating a PM's product analysis for cognitive gaps.
Research competitive positioning, recent moves, and market context for opportunity evaluation. Use during /pm:opportunity to understand the competitive dimension.
Analyzes growth loops, retention curves, and viral mechanics. Use when evaluating growth strategy, diagnosing retention problems, or validating growth model assumptions.
Searches past opportunities, compounds, and PM profile for similar past bets. Surfaces what worked, what failed, and what patterns exist. Use during opportunity evaluation.
North Star Metrics, ecosystem analysis, goal-setting, and tradeoff frameworks. Use when defining metrics, setting team goals, or resolving product tradeoffs.
Apply behavioral psychology, UX principles, and decision science to product design. Use when designing user flows, evaluating UX quality, or diagnosing adoption problems.
Growth loops, retention frameworks, viral mechanics, and acquisition models. Use when designing growth engines, diagnosing retention problems, or evaluating acquisition strategies.
Design complete metrics frameworks with outcome-intermediate-leading hierarchy, activation funnels, and engagement dimensions. Use when defining how to measure product success.
Evaluate product opportunities using the 5-question customer problem framework, business context analysis, and prioritization matrices. Use when assessing whether an idea is worth pursuing.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
No model invocation
Executes directly as bash, bypassing the AI model
No model invocation
Executes directly as bash, bypassing the AI model
A Claude Code plugin marketplace for product managers.
| Plugin | Description | Version |
|---|---|---|
| flywheel-pm | Small pushes, unstoppable momentum — 32 commands, 13 agents, 13 skills. Your PM operating system that gets sharper with every product cycle | 2.2.0 |
v1.0 introduced artifact compounding — specs, decisions, and learnings get better with every cycle. v2.0 adds a second layer: judgment compounding — simulation accuracy, bias awareness, and segmentation instincts get sharper the more you use them.
Two layers of compounding:
v2.0 adds 15 new commands, 5 new skills, and 4 new agents across skill development, strategy, and growth.
# Add the marketplace
claude plugin marketplace add https://github.com/abhitsian/flywheel-pm-marketplace
# Install the plugin
claude plugin install flywheel-pm
Restart Claude Code after installation.
/pm:welcome # Guided first-run experience
/pm:onboard # Create your PM profile
/pm:context # Load your work environment
Then run the core loop on something real:
/pm:opportunity # Frame the problem
/pm:solution # Design solutions
/pm:spec # Write the spec
/pm:review # 5 agents review in parallel
/pm:measure # Define predictions
/pm:compound # Capture learnings (never skip this)
Build judgment with the v2.0 skill-development commands:
/pm:teardown # Tear down a product to train simulation muscle
/pm:simulate # Simulate user reactions before shipping
/pm:bias-check # Check your decisions for cognitive biases
Flywheel PM encodes your PM operating system into three knowledge layers:
/pm:compound after each cycle./pm:share.Every command loads all three layers to produce personalized, context-aware output.
Observe → Opportunity → Solution → Spec → Review → Measure → Compound
^ |
└──────────────── learnings feed back ──────────────────────────┘
Teardown → Simulate → Bias-Check → Compound
^ |
└──────── judgment sharpens ─────────┘
Cycle 1: Generic frameworks. Predictions are guesses. Cycle 10: System knows your estimation bias, flags similar past failures, pre-fills metrics by project type. Cycle 20: Your second brain knows more about your PM operating system than you consciously remember.
| Command | Purpose |
|---|---|
/pm:welcome | Guided first-run experience |
/pm:onboard | Create your PM profile from docs or interactive Q&A |
/pm:context | Set up your operating environment |
| Command | Purpose |
|---|---|
/pm:research | Ingest signals from support, NPS, feature requests, sales, usage data |
/pm:interview | Prepare guides, capture notes, synthesize patterns |
| Command | Purpose |
|---|---|
/pm:opportunity | Signal -> problem -> business context -> prioritized opportunity |
/pm:solution | 20 creativity techniques -> risk/impact matrix -> classification |
/pm:spec | Vision narrative -> journey map -> metrics -> launch criteria |
/pm:review | 5 specialized agents review your spec in parallel |
/pm:measure | Predictions framework with activation funnel and continue/pause/pivot criteria |
/pm:compound | Compare predictions vs actuals, update profile, share with team |
| Command | Purpose |
|---|---|
/pm:decide | Frame decisions — options, criteria, pre-mortem, reversibility |
/pm:buyin | Stakeholder-specific narratives using 6 buy-in techniques |
/pm:product-review | Weekly reviews for team, leadership, or all-hands |
/pm:share | Promote a learning to the team knowledge base |
npx claudepluginhub abhitsian/compound-pm-marketplace --plugin flywheel-pmTrack and share what you've done with Claude Code — a local dashboard (tokens, streaks, hours saved, what's running now) plus screenshot-ready receipt cards for Slack and Teams.
The AI-native PM operating system — deep, framework-grounded PM skills with live MCP tool integrations, chained sub-agent workflows, and persistent user memory. Built for solo PMs and founding PMs who need an AI partner that actually knows their product.
Product discovery skills for PMs: ideation, experiments, assumption testing, feature prioritization, and customer interview synthesis.
The Product Manager's Operating System. AI-powered skills for every PM workflow: PRDs, prioritization, competitive intel, stakeholder updates, launch planning, and more.
Product Manager Skills Pack - 37 skills across 5 modules: demand insight, solution design, growth iteration, risk management, and product strategy
Agent-first PM toolkit with 9 specialist agents and 18 skills for solo developers and small teams
18 production-ready Claude Code skills for Product Managers. Discovery, build, measure, communicate.