PromptArchitect MCP

Instructions

Implementation Reference

• src/tools/generatePrompt.ts:31-85 (handler)

  Core handler function that implements the generate_prompt tool logic. Attempts to use the PromptArchitect API first, falls back to local template generation, and computes metadata.
  export async function generatePrompt(input: GeneratePromptInput): Promise<{ prompt: string; template: string; metadata: { estimatedTokens: number; wordCount: number; hasStructure: boolean; }; }> { const { idea, template = 'general', context, targetModel, workspaceContext } = input; logger.info('Generating prompt', { template, targetModel, ideaLength: idea.length, hasWorkspaceContext: !!workspaceContext }); let generatedPrompt: string = ''; // Use PromptArchitect API if (isApiClientAvailable()) { try { const response = await apiGeneratePrompt({ idea, template, context, targetModel, workspaceContext, }); generatedPrompt = response.prompt; logger.info('Generated via PromptArchitect API'); } catch (error) { logger.warn('API request failed, using fallback', { error: error instanceof Error ? error.message : 'Unknown error' }); } } // Fallback template generation if (!generatedPrompt) { generatedPrompt = createFallbackPrompt(idea, template, context); logger.warn('Using fallback prompt generation'); } // Calculate metadata const wordCount = generatedPrompt.split(/\s+/).length; const estimatedTokens = Math.ceil(wordCount * 1.3); // Rough estimate const hasStructure = /^#+\s|^\d+\.|^-\s|^\*\s/m.test(generatedPrompt); return { prompt: generatedPrompt, template, metadata: { estimatedTokens, wordCount, hasStructure, }, }; }
• src/tools/generatePrompt.ts:15-27 (schema)

  Zod schema defining the input parameters for the generate_prompt tool, used for validation.
  export const generatePromptSchema = z.object({ idea: z.string().min(1).describe('The user\'s raw prompt idea or concept'), template: z.enum(['coding', 'writing', 'research', 'analysis', 'factcheck', 'general']) .optional() .default('general') .describe('Template type to use for generation'), context: z.string().optional().describe('Additional context or constraints'), targetModel: z.enum(['gpt-4', 'claude', 'gemini', 'general']) .optional() .default('general') .describe('Target AI model to optimize for'), workspaceContext: z.string().optional().describe('Project context including file structure, tech stack, dependencies, and any relevant code snippets to ensure the generated prompt aligns with the project scope'), });
• src/server.ts:207-218 (registration)

  MCP server handler for incoming generate_prompt tool calls: parses arguments with schema and invokes the generatePrompt handler.
  case 'generate_prompt': { const input = generatePromptSchema.parse(args); const result = await generatePrompt(input); return { content: [ { type: 'text', text: JSON.stringify(result, null, 2), }, ], }; }
• src/server.ts:73-113 (registration)

  Tool metadata registration in ListTools response, defining name 'generate_prompt', description, and input schema for MCP discovery.
  { name: 'generate_prompt', description: `Transform a raw idea into a well-structured, actionable prompt optimized for AI assistants. Use this tool when you need to: • Create a new prompt from scratch • Structure a vague idea into a clear request • Generate role-specific prompts (coding, writing, research, etc.) Supports templates: coding (for programming tasks), writing (for content creation), research (for investigation), analysis (for data/business analysis), factcheck (for verification), general (versatile). IMPORTANT: When available, pass workspace context (file structure, package.json, tech stack) to generate prompts that align with the user's project.`, inputSchema: { type: 'object', properties: { idea: { type: 'string', description: 'The raw idea or concept to transform into a prompt. Can be brief or detailed.', }, template: { type: 'string', enum: ['coding', 'writing', 'research', 'analysis', 'factcheck', 'general'], description: 'Template type to use. Default: auto-detected from idea or "general".', }, context: { type: 'string', description: 'Additional context like domain, constraints, or preferences.', }, targetModel: { type: 'string', enum: ['gpt-4', 'claude', 'gemini', 'general'], description: 'Target AI model for optimization. Default: "general".', }, workspaceContext: { type: 'string', description: 'Project context to ensure the prompt aligns with the codebase. Include: file/folder structure, package.json dependencies, tech stack (React, Node, etc.), relevant code snippets, and the original user request. This helps generate prompts that comply with project conventions.', }, }, required: ['idea'], }, },
• src/utils/apiClient.ts:120-142 (helper)

  Helper function that makes HTTP request to PromptArchitect backend API for primary prompt generation logic (used by handler).
  export async function apiGeneratePrompt(params: { idea: string; template?: string; context?: string; targetModel?: string; workspaceContext?: string; }): Promise<GenerateResponse> { logger.info('Generating prompt via API', { template: params.template, ideaLength: params.idea.length, hasWorkspaceContext: !!params.workspaceContext }); const response = await apiRequest<GenerateResponse>('/generate', { idea: params.idea, template: params.template || 'general', context: params.context, targetModel: params.targetModel || 'gemini', workspaceContext: params.workspaceContext, }); return response; }