From flywheel-pm
Cognitive empathy engine — simulation, metacognition, and intellectual flexibility for product decisions. Use when modeling user behavior, checking assumptions, or developing product intuition.
npx claudepluginhub abhitsian/compound-pm-marketplace --plugin flywheel-pmThis skill uses the workspace's default tool permissions.
Provide the cognitive empathy toolkit that underlies great product judgment. Simulation is the capacity to mentally model how others think, feel, and act — the core skill that separates good PMs from great ones.
Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.
Records polished WebM UI demo videos of web apps using Playwright with cursor overlay, natural pacing, and three-phase scripting. Activates for demo, walkthrough, screen recording, or tutorial requests.
Delivers idiomatic Kotlin patterns for null safety, immutability, sealed classes, coroutines, Flows, extensions, DSL builders, and Gradle DSL. Use when writing, reviewing, refactoring, or designing Kotlin code.
Provide the cognitive empathy toolkit that underlies great product judgment. Simulation is the capacity to mentally model how others think, feel, and act — the core skill that separates good PMs from great ones.
Based on Shreyas Doshi's Product Sense framework.
The ability to watch your own thought process and catch biases in real-time.
Practices:
Product application:
The ability to change your mind when evidence warrants it.
The 5-Question Test (use when stuck on a position):
Product application:
The capacity to mentally envision how others think, feel, and act based on their unique context.
Techniques:
Null Hypotheses:
Products must overcome these defaults. Anything that doesn't is a weakness.
Every user action is governed by:
Nudge × (Motivation − Friction)^Satisfaction = Adoption
| Type | Question | Example |
|---|---|---|
| Functional | What tasks do I want to perform? | "I need to track expenses" |
| Social | How do I want to be perceived? | "I want my team to see me as organized" |
| Emotional | How do I want to feel? | "I want to feel in control of my finances" |
| Personal | What does this say about me? | "I'm the kind of person who plans ahead" |
| Type | Questions to Ask |
|---|---|
| Comprehension | Do I understand this? Is the next action clear? |
| Decision | How many choices? How complex? What's the cost of being wrong? |
| Action | How hard to start? How hard to finish? |
| Habit | How inconsistent with my current behavior? |
| Emotional | What do I stand to lose? How anxious does this make me? |
| Type | Source | Example |
|---|---|---|
| Intrinsic | Within the user | Personal motivation, habits, self-imposed rules |
| Extrinsic | From the product/environment | Notifications, social proof, defaults, urgency |
Critical alignment: Nudges must match the dominant motivation type. A functional nudge ("save 2 hours") won't move a socially-motivated user ("your team is using this").
| Bias | Product Impact | Design Response |
|---|---|---|
| Status quo | Users resist change | Make defaults good; minimize required changes |
| Loss aversion | Fear of losing > joy of gaining | Frame as "don't miss out" vs "you could get" |
| Social proof | People follow others | Show what others are doing |
| Anchoring | First number sticks | Set anchors intentionally |
| Endowment | People overvalue what they own | Get users to invest (personalize) early |
| Ambiguity aversion | Prefer known risk over unknown | Provide clear information, reduce uncertainty |
| Bias | PM Impact | Mitigation |
|---|---|---|
| Confirmation | Seeking supporting evidence | Actively seek disconfirmation |
| Mimetic desire | Copying competitors | "Would I do this if no one else was?" |
| Sunk cost | Continuing failing projects | "If starting fresh, would I choose this?" |
| Planning fallacy | Underestimating effort | Use reference class forecasting |
| Survivorship | Learning only from successes | Study failures equally |
This skill is loaded as context by:
/pm:teardown — for simulation and bias analysis/pm:simulate — for per-segment user modeling/pm:bias-check — for PM decision bias detection/pm:friction-log — for friction categorization/pm:motivation-map — for adoption equation analysisIt can also be invoked directly when any command needs to model user behavior or check cognitive biases.