om-ux-shape

Turn vague product and AI feature ideas into decided, implementation-ready directions.

1|Updated Apr 23, 2026
One-click install
npx skills add https://github.com/tkogut/agents-os-core --skill om-ux-shape-tkogut
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: om-ux-shape
Source: https://github.com/tkogut/agents-os-core/tree/main/global_skills/om-ux-shape
Command: npx skills add https://github.com/tkogut/agents-os-core --skill om-ux-shape-tkogut

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Product and AI feature ideas often arrive vague, overcomplicated, or unjustified, leading teams to build screens before deciding anything. This Skill converts ambiguity into a concrete product decision with evidence labels, a recommended direction, scoped states, and validation plans. ## Core Features & Use Cases - Three working modes: Shape a vague idea, Review an existing concept or flow, or produce a Handoff specification ready for engineering. - AI necessity gate: Decides whether AI belongs in a feature at all, then applies human-AI interaction guidelines, reward-function trade-offs, and failure-state design. - Evidence discipline: Separates known facts, inferences, assumptions, and unknowns, tags claims with evidence tiers, and defines the smallest test that could change the decision. - Use Case: A PM has a request to "add AI to onboarding". The Skill runs the necessity gate, recommends a rules-based checklist with one AI-assisted step, writes the actual screen copy and error states, and defines a Wizard-of-Oz test before build. ## Quick Start Use om-ux-shape to turn my rough feature idea into a decided direction with concrete screens, states, and a validation plan.

Frequently Asked Questions about om-ux-shape

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I turn a vague feature idea into a product decision?▼

Use Shape mode: state the decision, actor, and outcome, build an evidence ledger separating known facts from assumptions, diagnose the core obstacle, compare two or three mechanisms, and recommend one direction with the decisive trade-off and smallest validation test.

How to decide whether a feature should use AI?▼

Run the AI necessity gate: identify the user problem independent of AI, check whether rules, search, or templates solve it more cheaply, and assess failure cost and recovery. Choose explicitly between no AI, rules first, AI-assisted, AI-collaborative, or AI-automated.

What is the difference between Shape, Review, and Handoff modes?▼

Shape turns a vague idea into a decided direction. Review evaluates an existing concept or flow and returns a keep, simplify, rethink, or stop verdict. Handoff writes implementation-ready behavior, states, and acceptance criteria for engineering.

Does this skill write code or implement features?▼

No. It produces product decisions and specifications only. Its Handoff output is written to be consumed by implementing skills or engineers, and it explicitly touches no code and no tracker items.

When should I not use a full feature-shaping process?▼

Skip the full workflow for small, low-risk, reversible decisions. The skill scales depth to risk: a minor change gets a short recommendation with concrete behavior, while consequential or AI-driven work gets the complete evidence and validation treatment.