Skip to main content
Glama

suggest_improvements

Analyze HTML code to identify and propose design enhancements in areas like accessibility, performance, UX, aesthetics, and responsiveness, tailored to specified goals and target audience.

Instructions

Get AI-powered suggestions for design improvements

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
contextNoContext about the design goals
focusAreasNoAreas to focus improvements on
htmlYesHTML code to analyze
targetAudienceNoTarget audience for the design

Implementation Reference

  • src/index.ts:331-360 (registration)
    Registration of the 'suggest_improvements' tool in the TOOLS array, including input schema definition.
    { name: 'suggest_improvements', description: 'Get AI-powered suggestions for design improvements', inputSchema: { type: 'object', properties: { html: { type: 'string', description: 'HTML code to analyze' }, context: { type: 'string', description: 'Context about the design goals' }, targetAudience: { type: 'string', description: 'Target audience for the design' }, focusAreas: { type: 'array', items: { type: 'string', enum: ['accessibility', 'performance', 'ux', 'aesthetics', 'responsiveness'] }, description: 'Areas to focus improvements on' } }, required: ['html'] } },
  • Placeholder handler function for the suggest_improvements tool, imported and called from index.ts.
    export async function suggestImprovements(args: SuggestionOptions) { return { content: [ { type: 'text', text: `Generated suggestions for ${args.context || 'general'}\nFocus: ${args.focus || 'all'}` } ] }; }
  • TypeScript interface defining the input options for the suggest_improvements tool.
    export interface SuggestionOptions { html: string; context?: 'landing-page' | 'dashboard' | 'component' | 'form' | 'navigation'; focus?: 'accessibility' | 'performance' | 'design' | 'ux' | 'seo' | 'all'; priority?: 'critical' | 'important' | 'nice-to-have' | 'all'; framework?: 'react' | 'vue' | 'svelte' | 'html'; targetAudience?: string; businessGoals?: string[]; }
  • src/index.ts:447-448 (registration)
    Dispatch handler in the switch statement that routes calls to the suggestImprovements function.
    case 'suggest_improvements': return await suggestImprovements(args as unknown as SuggestionOptions);
  • Full AI-powered implementation of suggestImprovements using Gemini API, potentially intended as the real handler (not currently wired).
    export async function suggestImprovements(args: SuggestionsRequest) { try { const { html, context = '', targetAudience = '', focusAreas = ['accessibility', 'performance', 'ux', 'aesthetics', 'responsiveness'] } = args; if (isGeminiAvailable()) { const prompt = `Analyze this HTML code and provide intelligent design improvement suggestions: HTML Code: ${html} ${context ? `Context: ${context}` : ''} ${targetAudience ? `Target Audience: ${targetAudience}` : ''} Focus Areas: ${focusAreas.join(', ')} Please provide detailed, actionable suggestions for improvement in the following areas: ${focusAreas.includes('accessibility') ? '**Accessibility**: WCAG compliance, screen reader support, keyboard navigation, color contrast, semantic HTML' : ''} ${focusAreas.includes('performance') ? '**Performance**: CSS optimization, bundle size, render performance, Core Web Vitals' : ''} ${focusAreas.includes('ux') ? '**User Experience**: Usability, interaction patterns, user flow, intuitive design' : ''} ${focusAreas.includes('aesthetics') ? '**Visual Design**: Typography, color harmony, spacing, visual hierarchy, modern design trends' : ''} ${focusAreas.includes('responsiveness') ? '**Responsive Design**: Mobile-first approach, breakpoint strategy, cross-device compatibility' : ''} For each suggestion, provide: 1. Current issue or opportunity 2. Specific recommendation 3. Implementation details with Tailwind classes 4. Expected impact/benefit 5. Priority level (High/Medium/Low) Format as structured markdown with clear sections and actionable code examples.`; const suggestions = await callGemini(prompt); return { content: [ { type: 'text', text: `# AI-Powered Design Improvement Suggestions\n\n${suggestions}` } ] }; } else { // Fallback manual suggestions const manualSuggestions = generateManualSuggestions(html, focusAreas); return { content: [ { type: 'text', text: manualSuggestions } ] }; } } catch (error) { console.error('Suggestions error:', error); throw new Error(`Failed to generate suggestions: ${error instanceof Error ? error.message : 'Unknown error'}`); } }

Other Tools

Related Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Tai-DT/mcp-tailwind-gemini'

If you have feedback or need assistance with the MCP directory API, please join our Discord server