Claude skill
Design Critique Facilitator
Design Critique Facilitator is a Claude skill in the AI Skills Library Product Design plugin. Facilitate a structured design critique across 5 lenses (clarity, hierarchy, consistency, interaction & feedback, edge cases), starting from the designer's intent, concerns and what's already…
What it does
The description in the skill's SKILL.md, which Claude reads to decide when to use it:
Facilitate a structured design critique across 5 lenses (clarity, hierarchy, consistency, interaction & feedback, edge cases), starting from the designer's intent, concerns and what's already decided, and tagging each fix by severity and owner. Asks before critiquing and measures before praising. Use when presenting work for feedback, to get specific critique anchored to the designer's intent instead of vague opinions.
How to use it
- Design Critique Facilitator comes inside the AI Skills Library Product Design plugin, so you get it by installing that plugin, along with its 15 other skills. A plugin installs as one unit, with all its skills. The plugin's page has the install steps, and plugins are saved to your Claude account.
- Ask Claude for the task. When your request matches the skill's description, Claude loads it.
- To pick it yourself in Claude chat, type
/or click + and choose design-critique-facilitator. In Claude Code and Cowork, type:/product-design-skills:design-critique-facilitator
Source
- Plugin
- AI Skills Library Product Design
- SKILL.md
- View on GitHub
- Repo stars
- 2
- Repo last pushed
- Nov 11, 2026
- Checked
- Nov 16, 2026
More skills in AI Skills Library Product Design
- Design Scope NegotiatorSort a design scope into Ship / Fast-follow / Cut using effort vs. impact-if-removed, with the reason for each. Use when pushing back on scope creep or negotiating what gets cut before a deadline.
- Edge Case FinderStress-test a feature across 5 failure dimensions (input, permissions, concurrency, dependencies, user behavior) before it ships. Use before handing a spec to engineering, to catch what breaks.
- Experiment DesignPlan a product experiment - hypothesis, control and variant, duration, decision rule - and flag when traffic is too low for a reliable result. Use when planning an A/B test or any product experiment.
- Explain to 4 AudiencesTranslate one design decision into 4 audience-specific explanations (10-year-old, engineering, executive, end user) to pressure-test whether it's actually clear.
- Five Whys Root CauseDig past a stakeholder's feature request to the real underlying need, using 5 consecutive Why questions, then surface alternative solutions.
- Persona PretestSimulate how a specific user persona would react to a design or flow, surfacing likely confusion, objections, and drop-off points before real user testing.
- PRD First DraftWrite a PRD first draft that's clear enough for engineers to build from and concise enough for executives to read in 5 minutes. Use when starting to spec a feature or product.
- Product ContextWrite down the context the other skills keep asking for: a PRODUCT.md for the product (users, roles, references, vocabulary, where known issues live) and a brief per feature (goal, stage, what's…
- Success Metrics DefinitionDesign a metrics framework (primary metric, leading/lagging indicators, counter-metrics) that proves a feature solved the real problem, not just that it was used.
- User Interview SynthesisTurn messy interview notes from multiple users into ranked insights, frustrations, and unspoken needs. Use after collecting raw notes from user interviews.
- Acceptance CriteriaTurn a user story into Given/When/Then acceptance criteria covering the happy path, errors, and edge cases. Use to define "done" before engineering starts building.
- Assumption ReversalFlip your current design assumptions 180° and generate concrete ideas from the reversed perspective. Use when stuck on ideas or wanting to pressure-test the current design direction.
- Cross Industry StealFind 3 solutions from unrelated industries that solve a structurally similar problem, and translate the strongest one into a concrete idea.
- Data InterpretationInterpret metrics or experiment results by separating fact from explanation and ruling out alternative causes before recommending action.
- Decision RationaleDocument a design decision between 2 to 4 options with must-meet conditions, explicit criteria, scenario testing, and a confidence-rated recommendation.