magi
Generate custom videos from text prompts using the FAL Image/Video MCP Server. Specify duration, aspect ratio, and creative input for tailored video creation.
Instructions
Magi - Creative video generation
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| aspect_ratio | No | 16:9 | |
| duration | No | ||
| prompt | Yes | Text prompt for video generation |
Implementation Reference
- src/index.ts:627-675 (handler)Core handler function for the 'magi' text-to-video tool. Extracts parameters, configures FAL client, calls fal.subscribe on 'fal-ai/magi' endpoint, processes video output with download/data URL/auto-open, and returns formatted JSON content.private async handleTextToVideo(args: any, model: any) { const { prompt, duration = 5, aspect_ratio = '16:9' } = args; try { // Configure FAL client lazily with query config override configureFalClient(this.currentQueryConfig); const inputParams: any = { prompt }; if (duration) inputParams.duration = duration; if (aspect_ratio) inputParams.aspect_ratio = aspect_ratio; const result = await fal.subscribe(model.endpoint, { input: inputParams }); const videoData = result.data as FalVideoResult; const videoProcessed = await downloadAndProcessVideo(videoData.video.url, model.id); return { content: [ { type: 'text', text: JSON.stringify({ model: model.name, id: model.id, endpoint: model.endpoint, prompt, video: { url: videoData.video.url, localPath: videoProcessed.localPath, ...(videoProcessed.dataUrl && { dataUrl: videoProcessed.dataUrl }), width: videoData.video.width, height: videoData.video.height, }, 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:373-379 (schema)Dynamically generates the input schema for text-to-video tools like 'magi': requires 'prompt', optional 'duration' (1-30s) and 'aspect_ratio'.} else if (category === 'textToVideo') { baseSchema.inputSchema.properties = { prompt: { type: 'string', description: 'Text prompt for video generation' }, duration: { type: 'number', default: 5, minimum: 1, maximum: 30 }, aspect_ratio: { type: 'string', enum: ['16:9', '9:16', '1:1', '4:3', '3:4'], default: '16:9' }, }; baseSchema.inputSchema.required = ['prompt'];
- src/index.ts:114-114 (registration)Registers 'magi' in MODEL_REGISTRY.textToVideo with endpoint 'fal-ai/magi', enabling dynamic tool creation and dispatch.{ id: 'magi', endpoint: 'fal-ai/magi', name: 'Magi', description: 'Creative video generation' },
- src/index.ts:406-407 (registration)Registers the 'magi' tool schema dynamically in MCP listTools response by iterating textToVideo models.for (const model of MODEL_REGISTRY.imageToVideo) { tools.push(this.generateToolSchema(model, 'imageToVideo'));
- src/index.ts:467-482 (handler)Dispatch logic in CallToolRequestSchema handler: looks up model by name 'magi', checks textToVideo registry, invokes handleTextToVideo.const model = getModelById(name); if (!model) { throw new McpError( ErrorCode.MethodNotFound, `Unknown model: ${name}` ); } // 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); }