FAL Image/Video MCP Server

Instructions

Implementation Reference

• src/index.ts:107-107 (registration)

  The registration entry for the 'hidream' tool in the MODEL_REGISTRY.imageGeneration array. This defines the tool's ID, FAL endpoint, name, and description used for dynamic tool registration and execution.

  { id: 'hidream', endpoint: 'fal-ai/hidream-i1-full', name: 'HiDream I1', description: 'High-resolution image generation' },

• src/index.ts:357-373 (schema)

  Dynamically generates the input schema for image generation tools like 'hidream', including parameters such as prompt, image_size, num_images, and model-specific options.

  if (category === 'imageGeneration') {
    baseSchema.inputSchema.properties = {
      prompt: { type: 'string', description: 'Text prompt for image generation' },
      image_size: { type: 'string', enum: ['square_hd', 'square', 'portrait_4_3', 'portrait_16_9', 'landscape_4_3', 'landscape_16_9'], default: 'landscape_4_3' },
      num_images: { type: 'number', default: 1, minimum: 1, maximum: 4 },
    };
    baseSchema.inputSchema.required = ['prompt'];
    
    // Add model-specific parameters
    if (model.id.includes('flux') || model.id.includes('stable_diffusion')) {
      baseSchema.inputSchema.properties.num_inference_steps = { type: 'number', default: 25, minimum: 1, maximum: 50 };
      baseSchema.inputSchema.properties.guidance_scale = { type: 'number', default: 3.5, minimum: 1, maximum: 20 };
    }
    if (model.id.includes('stable_diffusion') || model.id === 'ideogram_v3') {
      baseSchema.inputSchema.properties.negative_prompt = { type: 'string', description: 'Negative prompt' };
    }
  } else if (category === 'textToVideo') {

• src/index.ts:475-482 (handler)

  Dispatch logic in the tool call handler that routes 'hidream' calls (found in imageGeneration registry) to the specific handleImageGeneration function.

  // Determine category and handle accordingly
  if (MODEL_REGISTRY.imageGeneration.find(m => m.id === name)) {
    return await this.handleImageGeneration(args, model);
  } else if (MODEL_REGISTRY.textToVideo.find(m => m.id === name)) {
    return await this.handleTextToVideo(args, model);
  } else if (MODEL_REGISTRY.imageToVideo.find(m => m.id === name)) {
    return await this.handleImageToVideo(args, model);
  }

• src/index.ts:495-563 (handler)

  Core handler implementation for 'hidream': extracts arguments, configures FAL client, calls fal.subscribe('fal-ai/hidream-i1-full'), processes image outputs (downloads, data URLs, auto-open), and returns formatted content.

  private async handleImageGeneration(args: any, model: any) {
    const {
      prompt,
      image_size = 'landscape_4_3',
      num_inference_steps = 25,
      guidance_scale = 3.5,
      num_images = 1,
      negative_prompt,
      safety_tolerance,
      raw,
    } = args;
  
    try {
      // Configure FAL client lazily with query config override
      configureFalClient(this.currentQueryConfig);
      const inputParams: any = { prompt };
      
      // Add common parameters
      if (image_size) inputParams.image_size = image_size;
      if (num_images > 1) inputParams.num_images = num_images;
      
      // Add model-specific parameters based on model capabilities
      if (model.id.includes('flux') || model.id.includes('stable_diffusion')) {
        if (num_inference_steps) inputParams.num_inference_steps = num_inference_steps;
        if (guidance_scale) inputParams.guidance_scale = guidance_scale;
      }
      if ((model.id.includes('stable_diffusion') || model.id === 'ideogram_v3') && negative_prompt) {
        inputParams.negative_prompt = negative_prompt;
      }
      if (model.id.includes('flux_pro') && safety_tolerance) {
        inputParams.safety_tolerance = safety_tolerance;
      }
      if (model.id === 'flux_pro_ultra' && raw !== undefined) {
        inputParams.raw = raw;
      }
  
      const result = await fal.subscribe(model.endpoint, { input: inputParams });
      const imageData = result.data as FalImageResult;
  
      const processedImages = await downloadAndProcessImages(imageData.images, model.id);
  
      return {
        content: [
          {
            type: 'text',
            text: JSON.stringify({
              model: model.name,
              id: model.id,
              endpoint: model.endpoint,
              prompt,
              images: processedImages,
              metadata: inputParams,
              download_path: DOWNLOAD_PATH,
              data_url_settings: {
                enabled: ENABLE_DATA_URLS,
                max_size_mb: Math.round(MAX_DATA_URL_SIZE / 1024 / 1024),
              },
              autoopen_settings: {
                enabled: AUTOOPEN,
                note: AUTOOPEN ? "Files automatically opened with default application" : "Auto-open disabled"
              },
            }, null, 2),
          },
        ],
      };
    } catch (error) {
      throw new Error(`${model.name} generation failed: ${error}`);
    }
  }

• src/index.ts:140-143 (helper)

  Helper function used to retrieve the model configuration (endpoint etc.) for 'hidream' by its ID during tool dispatch.

  function getModelById(id: string) {
    const allModels = getAllModels();
    return allModels.find(model => model.id === id);
  }