@studiomeyer/mcp-video 1.0.0

package/src/schemas/post-production.ts

@@ -0,0 +1,273 @@
+/**
+ * MCP Tool Schemas for Post-Production features
+ */
+
+export const postProductionSchemas = [
+  // --- Background Music ---
+  {
+    name: 'add_background_music',
+    description: 'Add background music to a video. Supports fade in/out, volume control, and automatic looping. Does NOT re-encode the video (fast). Output: MP4 with AAC audio.',
+    annotations: {
+      title: 'Add Background Music',
+      readOnlyHint: false,
+      destructiveHint: false,
+      openWorldHint: false,
+      idempotentHint: true,
+    },
+    inputSchema: {
+      type: 'object' as const,
+      properties: {
+        videoPath: {
+          type: 'string',
+          description: 'Path to the video file',
+        },
+        musicPath: {
+          type: 'string',
+          description: 'Path to the music file (mp3, wav, aac, ogg)',
+        },
+        outputPath: {
+          type: 'string',
+          description: 'Output path (default: adds -music suffix)',
+        },
+        musicVolume: {
+          type: 'number',
+          description: 'Music volume 0.0-1.0 (default: 0.25). Recommended: 0.15-0.3 for background.',
+        },
+        fadeInDuration: {
+          type: 'number',
+          description: 'Fade in duration in seconds (default: 2)',
+        },
+        fadeOutDuration: {
+          type: 'number',
+          description: 'Fade out duration in seconds (default: 3)',
+        },
+        loopMusic: {
+          type: 'boolean',
+          description: 'Loop music if shorter than video (default: true)',
+        },
+      },
+      required: ['videoPath', 'musicPath'],
+    },
+  },
+
+  // --- Concatenation ---
+  {
+    name: 'concatenate_videos',
+    description: 'Concatenate multiple video clips into one with cinematic transitions. Automatically normalizes resolution and framerate. Supports 30+ transition types: fade, dissolve, wipeleft, slideright, circleopen, radial, etc.',
+    annotations: {
+      title: 'Concatenate Videos',
+      readOnlyHint: false,
+      destructiveHint: false,
+      openWorldHint: false,
+      idempotentHint: true,
+    },
+    inputSchema: {
+      type: 'object' as const,
+      properties: {
+        clips: {
+          type: 'array',
+          items: {
+            type: 'object',
+            properties: {
+              path: { type: 'string', description: 'Path to video clip' },
+              trimStart: { type: 'number', description: 'Optional: start time in seconds' },
+              trimEnd: { type: 'number', description: 'Optional: end time in seconds' },
+            },
+            required: ['path'],
+          },
+          description: 'Video clips in order',
+        },
+        outputPath: {
+          type: 'string',
+          description: 'Output path',
+        },
+        transition: {
+          type: 'string',
+          enum: [
+            'fade', 'fadeblack', 'fadewhite', 'dissolve',
+            'wipeleft', 'wiperight', 'wipeup', 'wipedown',
+            'slideleft', 'slideright', 'slideup', 'slidedown',
+            'smoothleft', 'smoothright', 'circleopen', 'circleclose',
+            'radial', 'pixelize', 'diagtl', 'diagtr',
+          ],
+          description: 'Transition type between clips (default: fade)',
+        },
+        transitionDuration: {
+          type: 'number',
+          description: 'Transition duration in seconds (default: 1)',
+        },
+      },
+      required: ['clips', 'outputPath'],
+    },
+  },
+
+  // --- Intro/Outro Generator ---
+  {
+    name: 'generate_intro',
+    description: 'Generate a cinematic intro or outro clip with animated text. Creates a solid-color background with fade-in title and optional subtitle. Perfect for branding.',
+    annotations: {
+      title: 'Generate Intro/Outro Clip',
+      readOnlyHint: false,
+      destructiveHint: false,
+      openWorldHint: false,
+      idempotentHint: true,
+    },
+    inputSchema: {
+      type: 'object' as const,
+      properties: {
+        text: {
+          type: 'string',
+          description: 'Main title text',
+        },
+        subtitle: {
+          type: 'string',
+          description: 'Optional subtitle text',
+        },
+        duration: {
+          type: 'number',
+          description: 'Duration in seconds (default: 3)',
+        },
+        backgroundColor: {
+          type: 'string',
+          description: 'Background color hex (default: #0a0a0a)',
+        },
+        textColor: {
+          type: 'string',
+          description: 'Text color (default: white)',
+        },
+        outputPath: {
+          type: 'string',
+          description: 'Output path',
+        },
+      },
+      required: ['text', 'outputPath'],
+    },
+  },
+
+  // --- Social Media Format ---
+  {
+    name: 'convert_social_format',
+    description: 'Convert a video to a specific social media format. Handles aspect ratio conversion with crop, padding, or blur-background strategy. Supports: instagram-reel, instagram-feed, youtube, youtube-short, tiktok, linkedin.',
+    annotations: {
+      title: 'Convert to Social Media Format',
+      readOnlyHint: false,
+      destructiveHint: false,
+      openWorldHint: false,
+      idempotentHint: true,
+    },
+    inputSchema: {
+      type: 'object' as const,
+      properties: {
+        inputPath: {
+          type: 'string',
+          description: 'Path to input video',
+        },
+        outputPath: {
+          type: 'string',
+          description: 'Output path',
+        },
+        format: {
+          type: 'string',
+          enum: ['instagram-reel', 'instagram-feed', 'instagram-story', 'youtube', 'youtube-short', 'tiktok', 'linkedin-landscape', 'linkedin-square'],
+          description: 'Target social media format',
+        },
+        strategy: {
+          type: 'string',
+          enum: ['crop', 'pad', 'blur-background'],
+          description: 'How to handle aspect ratio: crop (cut edges), pad (black bars), blur-background (blurred fill). Default: blur-background.',
+        },
+      },
+      required: ['inputPath', 'outputPath', 'format'],
+    },
+  },
+
+  // --- Batch Social Format ---
+  {
+    name: 'convert_all_social_formats',
+    description: 'Convert a video to multiple social media formats at once. Creates separate files for each platform. Great for preparing content for all channels simultaneously.',
+    annotations: {
+      title: 'Convert to All Social Formats',
+      readOnlyHint: false,
+      destructiveHint: false,
+      openWorldHint: false,
+      idempotentHint: true,
+    },
+    inputSchema: {
+      type: 'object' as const,
+      properties: {
+        inputPath: {
+          type: 'string',
+          description: 'Path to input video',
+        },
+        outputDir: {
+          type: 'string',
+          description: 'Output directory (default: ./output/)',
+        },
+        formats: {
+          type: 'array',
+          items: {
+            type: 'string',
+            enum: ['instagram-reel', 'instagram-feed', 'instagram-story', 'youtube', 'youtube-short', 'tiktok', 'linkedin-landscape', 'linkedin-square'],
+          },
+          description: 'Formats to generate (default: instagram-reel, instagram-feed, youtube, tiktok)',
+        },
+        strategy: {
+          type: 'string',
+          enum: ['crop', 'pad', 'blur-background'],
+          description: 'Aspect ratio strategy (default: blur-background)',
+        },
+      },
+      required: ['inputPath'],
+    },
+  },
+
+  // --- Text Overlays ---
+  {
+    name: 'add_text_overlay',
+    description: 'Add animated text overlays to a video. Supports fade-in/out, positioning (center, top, bottom, corners), background boxes, and multiple text layers.',
+    annotations: {
+      title: 'Add Text Overlay to Video',
+      readOnlyHint: false,
+      destructiveHint: false,
+      openWorldHint: false,
+      idempotentHint: true,
+    },
+    inputSchema: {
+      type: 'object' as const,
+      properties: {
+        inputPath: {
+          type: 'string',
+          description: 'Path to input video',
+        },
+        outputPath: {
+          type: 'string',
+          description: 'Output path',
+        },
+        overlays: {
+          type: 'array',
+          items: {
+            type: 'object',
+            properties: {
+              text: { type: 'string', description: 'Text to display' },
+              position: {
+                type: 'string',
+                enum: ['center', 'top', 'bottom', 'top-left', 'top-right', 'bottom-left', 'bottom-right'],
+                description: 'Position (default: center)',
+              },
+              fontSize: { type: 'number', description: 'Font size (default: 48)' },
+              fontColor: { type: 'string', description: 'Color (default: white)' },
+              startTime: { type: 'number', description: 'Start time in seconds' },
+              endTime: { type: 'number', description: 'End time in seconds' },
+              fadeIn: { type: 'number', description: 'Fade in seconds (default: 0.5)' },
+              fadeOut: { type: 'number', description: 'Fade out seconds (default: 0.5)' },
+              showBackground: { type: 'boolean', description: 'Show background box (default: false)' },
+            },
+            required: ['text', 'startTime', 'endTime'],
+          },
+          description: 'Array of text overlays',
+        },
+      },
+      required: ['inputPath', 'outputPath', 'overlays'],
+    },
+  },
+];

package/src/schemas/smart-screenshot.ts

@@ -0,0 +1,122 @@
+/**
+ * MCP Tool Schema for Smart Screenshot Engine
+ */
+
+export const smartScreenshotSchemas = [
+  {
+    name: 'screenshot_element',
+    description: `Smart element-aware screenshot tool. Automatically detects and photographs specific features on a webpage.
+
+Instead of full-page screenshots, this finds and captures specific UI elements like:
+- Chat widgets, booking forms, pricing sections
+- Wizards/steppers, navigation, galleries
+- Contact forms, testimonials, CTAs, maps, video sections
+- Or ANY element via CSS selector
+
+Usage examples:
+  targets: ["chat"] → finds and screenshots the chat widget
+  targets: ["pricing", "booking"] → screenshots pricing section AND booking form
+  targets: ["all"] → auto-detects ALL features on the page
+  targets: [{ "feature": "hero", "selector": ".custom-hero" }] → custom selector
+  targets: [".my-section"] → direct CSS selector
+
+Supports 15+ feature types with intelligent DOM analysis.`,
+    annotations: {
+      title: 'Screenshot Page Element',
+      readOnlyHint: true,
+      destructiveHint: false,
+      openWorldHint: true,
+      idempotentHint: true,
+    },
+    inputSchema: {
+      type: 'object' as const,
+      properties: {
+        url: {
+          type: 'string',
+          description: 'URL of the website to screenshot',
+        },
+        targets: {
+          type: 'array',
+          items: {
+            oneOf: [
+              {
+                type: 'string',
+                description: 'Feature keyword (chat, pricing, booking, hero, nav, footer, gallery, wizard, testimonial, services, cta, map, video, newsletter, contact) or "all" for auto-detect, or CSS selector (.class, #id)',
+              },
+              {
+                type: 'object',
+                properties: {
+                  feature: { type: 'string', description: 'Feature name or keyword' },
+                  selector: { type: 'string', description: 'Custom CSS selector override' },
+                  padding: { type: 'number', description: 'Padding around element in px' },
+                  revealFirst: { type: 'boolean', description: 'Click to reveal element first (e.g. chat popup)' },
+                },
+                required: ['feature'],
+              },
+            ],
+          },
+          description: 'What to capture. Array of feature keywords, CSS selectors, or target objects. Use "all" to auto-detect everything.',
+        },
+        outputDir: {
+          type: 'string',
+          description: 'Output directory (default: ./output/smart-screenshots)',
+        },
+        viewport: {
+          type: 'object',
+          properties: {
+            width: { type: 'number', description: 'Viewport width (default: 1920)' },
+            height: { type: 'number', description: 'Viewport height (default: 1080)' },
+          },
+          description: 'Custom viewport dimensions',
+        },
+        darkMode: {
+          type: 'boolean',
+          description: 'Enable dark mode rendering (default: false)',
+        },
+        deviceScaleFactor: {
+          type: 'number',
+          description: 'Device scale factor for retina (default: 1, use 2 for retina)',
+        },
+        includeFullPage: {
+          type: 'boolean',
+          description: 'Also capture a full-page screenshot (default: false)',
+        },
+      },
+      required: ['url', 'targets'],
+    },
+  },
+  {
+    name: 'detect_page_features',
+    description: `Analyze a webpage and detect all available UI features/sections without taking screenshots.
+
+Returns a list of detected features with their names, locations, and sizes.
+Useful for planning which elements to screenshot or for understanding page structure.
+
+Detects: Hero, Chat, Booking, Pricing, Contact Form, Navigation, Footer, Gallery, Wizard/Stepper, Testimonials, Services, CTA, Map, Video, Newsletter sections.`,
+    annotations: {
+      title: 'Detect Page Features',
+      readOnlyHint: true,
+      destructiveHint: false,
+      openWorldHint: true,
+      idempotentHint: true,
+    },
+    inputSchema: {
+      type: 'object' as const,
+      properties: {
+        url: {
+          type: 'string',
+          description: 'URL of the website to analyze',
+        },
+        viewport: {
+          type: 'object',
+          properties: {
+            width: { type: 'number' },
+            height: { type: 'number' },
+          },
+          description: 'Viewport dimensions (default: 1920x1080)',
+        },
+      },
+      required: ['url'],
+    },
+  },
+];

package/src/schemas/tts.ts

@@ -0,0 +1,197 @@
+/**
+ * MCP Tool Schemas for TTS & Narrated Video
+ */
+
+const sceneSchema = {
+  type: 'object' as const,
+  properties: {
+    type: {
+      type: 'string',
+      enum: ['scroll', 'pause', 'hover', 'click', 'type', 'wait'],
+    },
+    to: { type: ['string', 'number'] },
+    duration: { type: 'number' },
+    easing: { type: 'string' },
+    selector: { type: 'string' },
+    text: { type: 'string' },
+    waitFor: { type: ['string', 'number'] },
+    pauseAfter: { type: 'number' },
+  },
+  required: ['type'],
+};
+
+export const ttsSchemas = [
+  // --- Text-to-Speech ---
+  {
+    name: 'generate_speech',
+    description: 'Generate speech audio from text using ElevenLabs (primary) or OpenAI TTS (fallback). Supports German, English, Spanish and 29+ languages. Returns an MP3 file. IMPORTANT: Uses paid APIs — ElevenLabs or OpenAI.',
+    annotations: {
+      title: 'Generate Speech (TTS)',
+      readOnlyHint: false,
+      destructiveHint: false,
+      openWorldHint: true,
+      idempotentHint: true,
+    },
+    inputSchema: {
+      type: 'object' as const,
+      properties: {
+        text: {
+          type: 'string',
+          description: 'Text to convert to speech',
+        },
+        outputPath: {
+          type: 'string',
+          description: 'Output path for MP3 file (default: ./output/speech-{timestamp}.mp3)',
+        },
+        provider: {
+          type: 'string',
+          enum: ['elevenlabs', 'openai'],
+          description: 'TTS provider (default: elevenlabs if API key available, otherwise openai)',
+        },
+        language: {
+          type: 'string',
+          description: 'Language code: en, de, es, fr, it, etc. (default: en)',
+        },
+        speed: {
+          type: 'number',
+          description: 'Speaking speed 0.7-1.2 for ElevenLabs, 0.25-4.0 for OpenAI (default: 1.0)',
+        },
+        elevenLabsVoice: {
+          type: 'string',
+          enum: ['rachel', 'sarah', 'emily', 'charlotte', 'alice', 'matilda', 'lily', 'brian', 'adam', 'daniel', 'josh', 'james', 'liam', 'chris', 'george'],
+          description: 'ElevenLabs voice (default: adam). Male: adam, brian, daniel, josh, james, liam. Female: rachel, sarah, emily, charlotte.',
+        },
+        elevenLabsModel: {
+          type: 'string',
+          enum: ['eleven_multilingual_v2', 'eleven_turbo_v2_5', 'eleven_flash_v2_5'],
+          description: 'ElevenLabs model (default: eleven_multilingual_v2 for best quality)',
+        },
+        openaiVoice: {
+          type: 'string',
+          enum: ['alloy', 'ash', 'coral', 'echo', 'fable', 'nova', 'onyx', 'sage', 'shimmer'],
+          description: 'OpenAI voice (default: onyx). Deep: onyx, echo. Bright: nova, shimmer. Neutral: alloy.',
+        },
+        openaiModel: {
+          type: 'string',
+          enum: ['tts-1', 'tts-1-hd'],
+          description: 'OpenAI model (default: tts-1-hd for best quality)',
+        },
+        stability: {
+          type: 'number',
+          description: 'ElevenLabs voice stability 0-1 (default: 0.5). Lower = more dynamic.',
+        },
+        similarityBoost: {
+          type: 'number',
+          description: 'ElevenLabs similarity 0-1 (default: 0.75).',
+        },
+      },
+      required: ['text'],
+    },
+  },
+
+  // --- List Voices ---
+  {
+    name: 'list_voices',
+    description: 'List all available ElevenLabs voices including custom/cloned voices. Requires ELEVENLABS_API_KEY.',
+    annotations: {
+      title: 'List ElevenLabs Voices',
+      readOnlyHint: true,
+      destructiveHint: false,
+      openWorldHint: true,
+      idempotentHint: true,
+    },
+    inputSchema: {
+      type: 'object' as const,
+      properties: {},
+      required: [],
+    },
+  },
+
+  // --- Narrated Video (the killer feature) ---
+  {
+    name: 'create_narrated_video',
+    description: `Create a fully narrated explainer video of a website or app. This is the ultimate tool: provide a URL and a script, and get a professional video with AI voiceover synchronized to the on-screen actions.
+
+Flow: Script → TTS → Website Recording (synced to speech durations) → Merge video + audio → Final MP4
+
+Each segment has text (what to say) and a scene (what to show). The scene duration is automatically matched to the speech duration.
+
+IMPORTANT: Uses paid TTS APIs (ElevenLabs or OpenAI). Always confirm with user before calling.`,
+    annotations: {
+      title: 'Create Narrated Video',
+      readOnlyHint: false,
+      destructiveHint: false,
+      openWorldHint: true,
+      idempotentHint: false,
+    },
+    inputSchema: {
+      type: 'object' as const,
+      properties: {
+        url: {
+          type: 'string',
+          description: 'URL of the website/app to record',
+        },
+        segments: {
+          type: 'array',
+          items: {
+            type: 'object',
+            properties: {
+              text: {
+                type: 'string',
+                description: 'Narration text for this segment',
+              },
+              scene: sceneSchema,
+              paddingAfter: {
+                type: 'number',
+                description: 'Extra pause after speech in seconds (default: 0.5)',
+              },
+            },
+            required: ['text', 'scene'],
+          },
+          description: 'Array of narration segments, each with text + scene action',
+        },
+        outputPath: {
+          type: 'string',
+          description: 'Output path without extension (default: ./output/narrated-{domain})',
+        },
+        provider: {
+          type: 'string',
+          enum: ['elevenlabs', 'openai'],
+          description: 'TTS provider (default: elevenlabs)',
+        },
+        language: {
+          type: 'string',
+          description: 'Language code (default: en)',
+        },
+        viewport: {
+          type: 'string',
+          enum: ['desktop', 'desktop-4k', 'tablet', 'mobile'],
+          description: 'Viewport (default: desktop)',
+        },
+        elevenLabsVoice: {
+          type: 'string',
+          enum: ['rachel', 'sarah', 'emily', 'charlotte', 'alice', 'matilda', 'lily', 'brian', 'adam', 'daniel', 'josh', 'james', 'liam', 'chris', 'george'],
+          description: 'ElevenLabs voice (default: adam)',
+        },
+        openaiVoice: {
+          type: 'string',
+          enum: ['alloy', 'ash', 'coral', 'echo', 'fable', 'nova', 'onyx', 'sage', 'shimmer'],
+          description: 'OpenAI voice (default: onyx)',
+        },
+        speed: {
+          type: 'number',
+          description: 'Speaking speed (default: 1.0)',
+        },
+        backgroundMusicPath: {
+          type: 'string',
+          description: 'Optional: path to background music file',
+        },
+        backgroundMusicVolume: {
+          type: 'number',
+          description: 'Background music volume 0-1 (default: 0.1)',
+        },
+      },
+      required: ['url', 'segments'],
+    },
+  },
+];