Prompts
22 expert methods. Each tells your assistant which tools to call and how to read a Figma link.
How to run one
- Claude Code: type
/mcp__designers__critiqueand fill in the arguments it asks for. - Claude Desktop and claude.ai: click the plus menu, choose Add from Designers, pick the prompt.
- Any other app: ask in words: Use the critique prompt from the Designers server on this Figma link: ...
| Prompt | What it does | Tier |
|---|---|---|
getting_started | Explains the server and suggests the three best first tasks for what you have (a Figma link, a live URL, some copy, a research plan, an event list). | Free |
critique | A structured critique of a design, framed by its stage (concept to shipped), ending in observations, impacts and a short list of changes. Refuses feedback that belongs to a later stage. | Free |
heuristic_evaluation | Walks one task through a flow and records findings against Nielsen's ten heuristics with 0 to 4 severity, using the worksheet tool for signals. You do the judging. | Free |
accessibility_review | Reviews a Figma frame, a live page or a description against WCAG 2.2 using exact contrast maths and a markup audit, and says what still needs a human check. | Free |
microcopy_review | Reviews buttons, errors, empty states and other interface text for clarity, tone and consistency, and removes the AI-writing tells. | Free |
slop_review | Finds what looks generated in a live page or a design (gradients, default fonts, eyebrows, card grids, stock copy) and says what to do instead. | Free |
research_plan | Turns a decision and a question into a research plan, picks the method, writes a screener and an interview or test guide, and lints the questions. | Free |
feedback_synthesis | Turns a pile of feedback (interviews, tickets, survey comments, critique notes) into evidence-backed themes, severity and next actions. | Free |
tracking_plan | Turns a flow or a product into an analytics tracking plan with named events, properties, funnels and the questions each answers, then lints it. | Free |
brief_to_ui | Turns a product idea into a design brief a model can build from: point of view, type, colour, layout, content and constraints, so the result is not the generic template. | Free |
handoff_spec | Produces a handoff document from a Figma frame or a design: behaviour, states, tokens, content rules, accessibility, edge cases and acceptance criteria. | Free |
token_audit | Checks colour, type and spacing tokens for naming, duplication, scale discipline and contrast, using the exact values from the file. | Free |
design_system_audit | A full design system audit across foundations, tokens, components, accessibility, adoption and governance, with a scorecard and a 30-60-90 day plan. | Free |
design_qa | Compares a built page or screenshots to the Figma design and reports differences that matter to people, with measured values and an accessibility check. | Free |
art_direction_brief | Writes image prompts and a style guide that produce a coherent set of AI-generated images with a point of view, avoiding the stock-illustration and glowing-blob look. | Free |
stakeholder_presentation | Builds the story for presenting design work to stakeholders: the decision, the evidence, the options, and the ask, with the objections to expect. | Free |
industry_playbook | Reviews a design against a playbook for a specific product type: e-commerce checkout, SaaS onboarding, forms, dashboards or mobile. | Pro |
start_here | The router. Asks what you have and what you want, then points you to the right tool or prompt. Same as getting_started. | Free |
troubleshoot | Walks through connection and setup problems with the doctor tool and the troubleshooting tree. Use when something is connected but not working, or you hit 401, 403 or limits. | Free |
design_md | Interviews you or reads an existing site or Figma file, then writes a DESIGN.md (the context file models read before building) and checks it. | Free |
critique_compare | Compares two options or a before and after against a goal, forcing a pick, real problems in both, and evidence. Counters the tendency of models to praise whatever they see. | Free |
microcopy_write | Writes the small text a product needs (empty states, errors, confirmations, tooltips, onboarding hints) in the product's voice, then checks it. | Free |