video-context-mcp-server 0.11.0-beta

package/README.md

@@ -0,0 +1,432 @@
+# Video Context MCP Server
+
+Video Context MCP Server is a Model Context Protocol (MCP) server that gives MCP-compatible coding assistants (such as GitHub Copilot in VS Code, Cursor, and Claude Code) the ability to understand and analyze video content.
+
+## Features
+
+- 🎬 **Video Q&A** — Ask questions about video content and get AI-powered answers
+- 📝 **Video Summarization** — Generate structured summaries with key scenes and timelines
+- 🖼️ **Frame Extraction** — Extract frames at specific timestamps or intervals
+- 🔍 **Timestamp Search** — Find the exact moment when something happens in a video
+- 📊 **Video Metadata** — Get duration, resolution, fps, codec, and other technical details
+- 🎙️ **Audio Transcription** — Transcribe speech from any video using Deepgram, AssemblyAI, Groq/Whisper, or Gemini
+- 🔊 **Speaker Diarization** — Identify who said what (Deepgram and AssemblyAI)
+- 🔄 **Multi-Backend Support** — Choose between Gemini (native multimodal), GLM-4.6V (cheap/free), or Kimi K2.5 (broader format support)
+- 🎯 **Smart Video Handling** — Extracts keyframes from long videos to reduce token usage (when not using Gemini)
+
+## Installation
+
+### Quick Start (Recommended for ordinary users)
+
+#### 1. Prerequisites
+
+- Node.js 18+
+- VS Code with GitHub Copilot Chat enabled
+
+#### 2. Get API keys
+
+You'll need API keys for one or more **video** backends:
+
+- **Gemini 3 Flash Preview (Google)**: [Get API Key](https://aistudio.google.com/app/apikey)
+- **Kimi K2.5 (Moonshot AI)**: [Get API Key](https://platform.moonshot.ai)
+- **GLM-4.6V (Z.AI)**: [Get API Key](https://z.ai/manage-apikey/apikey-list)
+
+For **audio transcription** (`transcribe_video`), you'll also need at least one audio provider key:
+
+- **Deepgram** (default): [Get API Key](https://console.deepgram.com/)
+- **AssemblyAI**: [Get API Key](https://www.assemblyai.com/dashboard)
+- **Groq** (free Whisper): [Get API Key](https://console.groq.com/)
+- **Gemini** (reuse `GEMINI_API_KEY` above — no extra key needed)
+
+#### 3. Install the MCP server
+
+```bash
+npm install -g video-context-mcp-server
+```
+
+This installs the executable command: `video-context-mcp`.
+
+> **Tip:** Periodically re-run the above command to get the latest version:
+>
+> ```bash
+> npm install -g video-context-mcp-server@latest
+> ```
+
+#### 4. Configure VS Code MCP
+
+Create (or update) `.vscode/mcp.json` in your project/workspace:
+
+```json
+{
+  "servers": {
+    "videoMcp": {
+      "type": "stdio",
+      "command": "video-context-mcp",
+      "env": {
+        "GEMINI_API_KEY": "your-gemini-key",
+        "MOONSHOT_API_KEY": "your-moonshot-key",
+        "Z_AI_API_KEY": "your-zai-key",
+        "DEEPGRAM_API_KEY": "your-deepgram-key",
+        "ASSEMBLYAI_API_KEY": "your-assemblyai-key",
+        "GROQ_API_KEY": "your-groq-key"
+      }
+    }
+  }
+}
+```
+
+Open Copilot Chat in VS Code. The MCP server starts automatically when tools are needed.
+
+### Configure Cursor MCP
+
+Add this server to your Cursor MCP configuration (global or project-level):
+
+```json
+{
+  "mcpServers": {
+    "videoMcp": {
+      "command": "video-context-mcp",
+      "env": {
+        "GEMINI_API_KEY": "your-gemini-key",
+        "MOONSHOT_API_KEY": "your-moonshot-key",
+        "Z_AI_API_KEY": "your-zai-key",
+        "DEEPGRAM_API_KEY": "your-deepgram-key",
+        "ASSEMBLYAI_API_KEY": "your-assemblyai-key",
+        "GROQ_API_KEY": "your-groq-key"
+      }
+    }
+  }
+}
+```
+
+Notes:
+
+- Run `npm install -g video-context-mcp-server` first if you haven't already.
+- If you prefer not to install globally, use `npx -y video-context-mcp-server@latest` as the command (slower startup due to registry check).
+- Set one or both API keys depending on which provider you use.
+
+### Configure Claude Code MCP
+
+Use the Claude CLI to register the MCP server:
+
+```bash
+claude mcp add videoMcp video-context-mcp \
+  --env GEMINI_API_KEY=your-gemini-key \
+  --env MOONSHOT_API_KEY=your-moonshot-key \
+  --env Z_AI_API_KEY=your-zai-key \
+  --env DEEPGRAM_API_KEY=your-deepgram-key \
+  --env ASSEMBLYAI_API_KEY=your-assemblyai-key \
+  --env GROQ_API_KEY=your-groq-key
+```
+
+Then verify with:
+
+```bash
+claude mcp list
+```
+
+If you prefer not to install globally, register via `npx` instead (slower startup due to registry check):
+
+```bash
+claude mcp add videoMcp npx -y video-context-mcp-server@latest \
+  --env MOONSHOT_API_KEY=your-moonshot-key \
+  --env Z_AI_API_KEY=your-zai-key \
+  --env DEEPGRAM_API_KEY=your-deepgram-key \
+  --env ASSEMBLYAI_API_KEY=your-assemblyai-key \
+  --env GROQ_API_KEY=your-groq-key
+```
+
+### Troubleshooting Setup
+
+- **`video-context-mcp: command not found`**
+  - Make sure Node.js is installed and available in your shell (`node -v`, `npm -v`).
+  - If installed globally, re-run: `npm install -g video-context-mcp-server`.
+  - If global binaries are not on PATH, use `npx -y video-context-mcp-server@latest` instead of `video-context-mcp`.
+
+- **MCP server not appearing in client**
+  - Restart the client app after changing MCP configuration.
+  - Validate JSON syntax in your MCP config file.
+  - For Claude Code, verify registration with `claude mcp list`.
+
+- **Missing API key errors**
+  - Set `GEMINI_API_KEY` for Gemini usage, `Z_AI_API_KEY` for GLM usage, and `MOONSHOT_API_KEY` for Kimi usage.
+  - For audio transcription, set `DEEPGRAM_API_KEY` (default), `ASSEMBLYAI_API_KEY`, `GROQ_API_KEY` (free tier), or reuse `GEMINI_API_KEY`. At least one audio key is needed to use `transcribe_video`.
+  - You can set only the keys for the providers you intend to use.
+  - For local files, if `provider=glm` is requested but `Z_AI_API_KEY` is missing, the server automatically falls back to Kimi (and then Gemini) when those keys are available.
+  - For remote `http(s)` video URLs, all three AI providers (`glm`, `kimi`, `gemini`) are supported. All tools automatically download remote videos to a temporary file before processing.
+
+### Alternative: Run via npx without global install
+
+If you prefer not to install globally, you can use `npx` instead. Note this adds a startup delay due to the npm registry check on each run:
+
+```json
+{
+  "servers": {
+    "videoMcp": {
+      "type": "stdio",
+      "command": "npx",
+      "args": ["-y", "video-context-mcp-server@latest"],
+      "env": {
+        "GEMINI_API_KEY": "your-gemini-key",
+        "MOONSHOT_API_KEY": "your-moonshot-key",
+        "Z_AI_API_KEY": "your-zai-key",
+        "DEEPGRAM_API_KEY": "your-deepgram-key",
+        "ASSEMBLYAI_API_KEY": "your-assemblyai-key",
+        "GROQ_API_KEY": "your-groq-key"
+      }
+    }
+  }
+}
+```
+
+The `@latest` tag ensures you always get the newest published version, at the cost of a network round-trip on every startup — this also means the `npx` approach self-updates automatically and requires no manual update step.
+
+### For contributors: install from source
+
+Clone this repository, then:
+
+```bash
+npm install
+npm run build
+```
+
+Then use this `.vscode/mcp.json` server command:
+
+```json
+{
+  "servers": {
+    "videoMcp": {
+      "type": "stdio",
+      "command": "node",
+      "args": ["${workspaceFolder}/dist/index.js"],
+      "env": {
+        "GEMINI_API_KEY": "your-gemini-key",
+        "MOONSHOT_API_KEY": "your-moonshot-key",
+        "Z_AI_API_KEY": "your-zai-key",
+        "DEEPGRAM_API_KEY": "your-deepgram-key",
+        "ASSEMBLYAI_API_KEY": "your-assemblyai-key",
+        "GROQ_API_KEY": "your-groq-key"
+      }
+    }
+  }
+}
+```
+
+### Debugging Behavior in VS Code
+
+When your `.vscode/mcp.json` includes a `dev` block such as:
+
+```jsonc
+{
+  "servers": {
+    "videoMcp": {
+      "type": "stdio",
+      "command": "node",
+      "args": ["${workspaceFolder}/dist/index.js"],
+      "env": {
+        "GEMINI_API_KEY": "your-gemini-key",
+        "MOONSHOT_API_KEY": "your-moonshot-key",
+        "Z_AI_API_KEY": "your-zai-key",
+        "DEEPGRAM_API_KEY": "your-deepgram-key",
+        "ASSEMBLYAI_API_KEY": "your-assemblyai-key",
+        "GROQ_API_KEY": "your-groq-key",
+      },
+      "dev": {
+        "watch": "src/**/*.ts",
+        "debug": { "type": "node" },
+      },
+    },
+  },
+}
+```
+
+you may see frequent logs in the **Output** panel under `MCP: videoMcp`.
+
+This is expected in development mode:
+
+- `watch` restarts/reloads the MCP server when TypeScript files change
+- `debug` enables Node debug integration
+- MCP protocol payloads (tool schemas, discovery events, lifecycle messages) are printed in that output channel
+
+If you want less noise, remove either:
+
+- `dev.debug` (keeps auto-watch, disables debug integration), or
+- the full `dev` block (disables watch + debug behavior)
+
+## Available Tools
+
+| Tool               | Description                          | Parameters                                                      |
+| ------------------ | ------------------------------------ | --------------------------------------------------------------- |
+| `analyze_video`    | Ask questions about video content    | `videoPath`, `question`, `provider?`                            |
+| `summarize_video`  | Generate a structured video summary  | `videoPath`, `provider?`                                        |
+| `extract_frames`   | Extract frames from a video          | `videoPath`, `mode`, `count/intervalSec/timestamps`             |
+| `search_timestamp` | Find when something specific happens | `videoPath`, `query`, `provider?`                               |
+| `get_video_info`   | Get video metadata                   | `videoPath`                                                     |
+| `transcribe_video` | Transcribe audio/speech from a video | `videoPath`, `provider?`, `language?`, `diarize?`, `translate?` |
+
+### Path and Provider Constraints
+
+- **All 6 tools** support both local files and remote `http(s)` URLs. Remote videos are automatically downloaded to a temporary file before processing.
+- For remote `http(s)` URLs with AI-powered video tools (`analyze_video`, `summarize_video`, `search_timestamp`): all three providers (`provider=gemini`, `provider=glm`, `provider=kimi`) are supported. Remote videos are downloaded to a temp file before upload.
+- For `transcribe_video`, all four audio providers (Deepgram, AssemblyAI, Groq, Gemini) support both local files and remote URLs.
+- For local inputs, all tools accept normal filesystem paths or `file://` URIs (automatically normalized).
+
+## Usage Examples
+
+### Analyze Video
+
+Ask Copilot Chat:
+
+> "Analyze the video at `./demo.mp4` and tell me what happens in it"
+
+### Summarize Video
+
+> "Summarize the video at `./long-video.mp4`"
+
+### Extract Frames
+
+> "Extract 5 evenly-spaced frames from `./video.mp4`"
+
+> "Extract a frame at timestamp 30 seconds from `./video.mp4`"
+
+### Search Timestamp
+
+> "In `./video.mp4`, at what timestamp does the person wave?"
+
+### Get Video Info
+
+> "Get the video info for `./video.mp4`"
+
+### Transcribe Video
+
+> "Transcribe the audio from `./meeting.mp4`"
+
+> "Transcribe `./interview.mp4` with speaker diarization using AssemblyAI"
+
+> "Transcribe this Spanish video and translate it to English: `./video.mp4`"
+
+## Additional Guides
+
+- [Screen Recording for Small File Sizes (Windows)](docs/screen-recording-small-files.md)
+
+## Backend Comparison
+
+| Feature        | Gemini 3 Flash Preview                         | GLM-4.6V                      | Kimi K2.5                                      |
+| -------------- | ---------------------------------------------- | ----------------------------- | ---------------------------------------------- |
+| Video formats  | mp4, mpeg, mov, avi, flv, mpg, webm, wmv, 3gpp | mp4, avi, mov, wmv, webm, m4v | mp4, mpeg, mov, avi, flv, mpg, webm, wmv, 3gpp |
+| Price          | Free tier available                            | $0.30 input / $0.90 output    | $0.60 input / $3.00 output                     |
+| Free tier      | Yes                                            | Yes (GLM-4.6V-Flash)          | No                                             |
+| Context window | 1M tokens                                      | 128K                          | 256K                                           |
+| Max file size  | 2 GB                                           | ~20 MB (base64)               | 100 MB                                         |
+| Best for       | Fallback only (inaccurate despite features)    | **Default** (free tier)       | Alternative to GLM                             |
+
+**GLM-4.6V is the default backend** — it offers a free tier (GLM-4.6V-Flash), making it a good zero-cost starting point. Kimi K2.5 is a paid alternative with broader format support; accuracy between the two has not been systematically compared. Gemini 3 Flash Preview is used as the **last resort fallback** despite its superior technical features (1M token context, 2GB file size, native multimodal audio+video support) because it has **proven inaccurate for video content analysis** in practice. Set `VIDEO_MCP_DEFAULT_PROVIDER=kimi` or `VIDEO_MCP_DEFAULT_PROVIDER=gemini` to switch the default. This env default is used when a tool call omits the `provider` parameter.
+
+When a provider's API key is missing, the tool automatically falls back to the next available provider in the ranking chain (**GLM → Kimi → Gemini**) and includes a notice in the response, e.g. `Provider used: kimi (fell back from glm)`.
+
+## Environment Variables
+
+| Variable                       | Description                                                                                                                                                                                                                                                                                                                                                                                                                   | Required                                    |
+| ------------------------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------- |
+| `GEMINI_API_KEY`               | Google API key for Gemini 3 Flash Preview                                                                                                                                                                                                                                                                                                                                                                                     | Optional (required if using Gemini)         |
+| `MOONSHOT_API_KEY`             | Moonshot AI API key for Kimi K2.5                                                                                                                                                                                                                                                                                                                                                                                             | Optional (required if using Kimi)           |
+| `Z_AI_API_KEY`                 | Z.AI API key for GLM-4.6V                                                                                                                                                                                                                                                                                                                                                                                                     | Optional (required if using GLM)            |
+| `VIDEO_MCP_DEFAULT_PROVIDER`   | Default video backend (`gemini`, `glm`, `kimi`)                                                                                                                                                                                                                                                                                                                                                                               | Optional (default: `glm`)                   |
+| `VIDEO_MCP_MAX_FRAMES`         | Max frames for summarization (GLM/Kimi only)                                                                                                                                                                                                                                                                                                                                                                                  | Optional (default: 20; clamped to 5-100)    |
+| `DEEPGRAM_API_KEY`             | Deepgram API key for `transcribe_video`                                                                                                                                                                                                                                                                                                                                                                                       | Optional (required if using Deepgram)       |
+| `ASSEMBLYAI_API_KEY`           | AssemblyAI API key for `transcribe_video`                                                                                                                                                                                                                                                                                                                                                                                     | Optional (required if using AssemblyAI)     |
+| `GROQ_API_KEY`                 | Groq API key for Whisper transcription via `transcribe_video`                                                                                                                                                                                                                                                                                                                                                                 | Optional (required if using Groq)           |
+| `AUDIO_MCP_DEFAULT_PROVIDER`   | Default audio provider; defaults to `deepgram`. Falls back in order: **deepgram → assemblyai → groq → gemini** when the selected provider's key is unavailable. A fallback notice is included in the response.                                                                                                                                                                                                                | Optional (auto-selects from available keys) |
+| `AUDIO_ENHANCE_VIDEO_ANALYSIS` | Controls audio transcript injection into GLM/Kimi `analyze_video`/`summarize_video` prompts. `auto` (default) — transcribes only when the video has a detected audio track; `true` — always attempt transcription; `false` — disabled. A confidence label (`high`/`medium`/`low`) is included in the injected header so the model can weight the transcript appropriately. Gemini is always skipped (handles audio natively). | Optional (default: `auto`)                  |
+
+### Example Configuration
+
+```json
+{
+  "servers": {
+    "videoMcp": {
+      "type": "stdio",
+      "command": "video-context-mcp",
+      "env": {
+        "GEMINI_API_KEY": "your-gemini-key",
+        "Z_AI_API_KEY": "your-zai-key",
+        "MOONSHOT_API_KEY": "your-moonshot-key",
+        "DEEPGRAM_API_KEY": "your-deepgram-key",
+        "ASSEMBLYAI_API_KEY": "your-assemblyai-key",
+        "GROQ_API_KEY": "your-groq-key",
+        "VIDEO_MCP_DEFAULT_PROVIDER": "glm",
+        "AUDIO_ENHANCE_VIDEO_ANALYSIS": "auto"
+      }
+    }
+  }
+}
+```
+
+> **Note:** `VIDEO_MCP_MAX_FRAMES` only applies when using **GLM or Kimi** as the provider. Gemini uploads the full video natively and ignores this setting. Add it to `env` only if you are running with `VIDEO_MCP_DEFAULT_PROVIDER=gemini`.
+
+## Development
+
+```bash
+# Install dependencies
+npm install
+
+# Run in development (auto-restart on changes)
+npm run dev
+
+# Build for production
+npm run build
+
+# Run type checking
+npm run type-check
+
+# Run linter
+npm run lint
+
+# Run automated tests
+npm run test
+
+# Run tests in watch mode
+npm run test:watch
+
+# Run tests with coverage
+npm run test:coverage
+
+# Format all files
+npm run format
+
+# Check formatting only
+npm run format:check
+
+# Lint + Type-check + Format + Build
+npm run ltfb
+```
+
+## Architecture
+
+```
+video-mcp/
+├── src/
+│   ├── index.ts              # MCP server entry point
+│   ├── tools/                # MCP tool implementations
+│   │   └── transcribeVideo.ts  # Audio transcription tool
+│   ├── services/             # Backend clients (Kimi, GLM, Gemini, ffmpeg)
+│   │   └── audio/            # Audio provider clients (Deepgram, AssemblyAI, Groq)
+│   └── utils/                # Helpers (temp files, base64, audio injection)
+├── .vscode/
+│   └── mcp.json              # VS Code MCP configuration
+├── docs/
+│   └── technical/            # Technical documentation
+└── .github/
+    └── copilot-instructions.md  # Copilot AI assistant guidelines
+```
+
+## License
+
+MIT
+
+## Credits
+
+- [MCP SDK](https://github.com/modelcontextprotocol/typescript-sdk) by Anthropic
+- [Kimi K2.5](https://github.com/MoonshotAI/Kimi-K2.5) by Moonshot AI
+- [GLM-4.6V](https://docs.z.ai/guides/vlm/glm-4.6v) by Z.AI
+- [ffmpeg](https://ffmpeg.org/) for video processing

package/dist/index.d.ts

@@ -0,0 +1,3 @@
+#!/usr/bin/env node
+export {};
+//# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":""}

package/dist/index.js

@@ -0,0 +1,152 @@
+#!/usr/bin/env node
+import { McpServer } from '@modelcontextprotocol/sdk/server/mcp.js';
+import { StdioServerTransport } from '@modelcontextprotocol/sdk/server/stdio.js';
+import { z } from 'zod';
+// Import tool handlers
+import { analyzeVideoTool } from './tools/analyzeVideo.js';
+import { summarizeVideoTool } from './tools/summarizeVideo.js';
+import { extractFramesTool } from './tools/extractFrames.js';
+import { searchTimestampTool } from './tools/searchTimestamp.js';
+import { getVideoInfoTool } from './tools/getVideoInfo.js';
+import { transcribeVideoTool } from './tools/transcribeVideo.js';
+import { setLoggerServer } from './utils/logger.js';
+/**
+ * Main entry point for the Video Context MCP Server
+ * Creates an MCP server, registers all video analysis tools, and connects via stdio
+ */
+async function main() {
+    // Create MCP server with name and version
+    const server = new McpServer({
+        name: 'video-mcp',
+        version: '1.0.0',
+    }, {
+        capabilities: {
+            logging: {}, // Enable logging for progress reporting
+        },
+    });
+    // Wire up the logger so tools can emit progress notifications
+    setLoggerServer(server);
+    // Register all video analysis tools
+    // Tool 1: analyze_video - Ask questions about video content
+    server.registerTool('analyze_video', {
+        title: 'Analyze Video',
+        description: 'Ask questions about video content and get AI-powered answers. Supports both local files and URLs.',
+        inputSchema: z.object({
+            videoPath: z
+                .string()
+                .describe('Path to the video file (local path or URL)'),
+            question: z
+                .string()
+                .describe('Question to ask about the video content'),
+            provider: z
+                .enum(['glm', 'kimi', 'gemini'])
+                .optional()
+                .describe("AI backend to use: 'glm' (GLM-4.6V, default), 'kimi' (Kimi K2.5), or 'gemini' (Gemini 3 Flash Preview)"),
+        }),
+    }, analyzeVideoTool);
+    // Tool 2: summarize_video - Generate structured video summary
+    server.registerTool('summarize_video', {
+        title: 'Summarize Video',
+        description: 'Generate a structured summary of the video including overview, key scenes, and timeline. For long videos (>5 min), extracts keyframes to reduce token usage (unless using Gemini, which processes natively).',
+        inputSchema: z.object({
+            videoPath: z
+                .string()
+                .describe('Path to the video file (local path or URL)'),
+            provider: z
+                .enum(['glm', 'kimi', 'gemini'])
+                .optional()
+                .describe("AI backend to use: 'glm' (GLM-4.6V, default), 'kimi' (Kimi K2.5), or 'gemini' (Gemini 3 Flash Preview)"),
+        }),
+    }, summarizeVideoTool);
+    // Tool 3: extract_frames - Extract frames from video
+    server.registerTool('extract_frames', {
+        title: 'Extract Frames',
+        description: 'Extract frames from a video at specific timestamps or intervals. Supports local files (including file:// URIs) and remote http(s) URLs. No AI backend required.',
+        inputSchema: z.object({
+            videoPath: z
+                .string()
+                .describe('Path to the video file (local path, file:// URI, or http(s) URL)'),
+            mode: z
+                .enum(['even', 'interval', 'timestamps'])
+                .describe("Extraction mode: 'even' (N evenly-spaced frames), 'interval' (every N seconds), or 'timestamps' (at specific times)"),
+            count: z
+                .number()
+                .int()
+                .min(1)
+                .max(100)
+                .optional()
+                .describe("Number of frames to extract (required for 'even' mode)"),
+            intervalSec: z
+                .number()
+                .min(0.1)
+                .optional()
+                .describe("Interval in seconds between frames (required for 'interval' mode)"),
+            timestamps: z
+                .array(z.number().min(0))
+                .optional()
+                .describe("Array of timestamps in seconds (required for 'timestamps' mode)"),
+        }),
+    }, extractFramesTool);
+    // Tool 4: search_timestamp - Find when something happens in video
+    server.registerTool('search_timestamp', {
+        title: 'Search Timestamp',
+        description: 'Find the timestamp when something specific happens in a video. Extracts frames and uses AI to locate the content. Supports local files (including file:// URIs) and remote http(s) URLs.',
+        inputSchema: z.object({
+            videoPath: z
+                .string()
+                .describe('Path to the video file (local path, file:// URI, or http(s) URL)'),
+            query: z
+                .string()
+                .describe("What to search for, e.g., 'person waves', 'dog runs', 'car crash'"),
+            provider: z
+                .enum(['glm', 'kimi', 'gemini'])
+                .optional()
+                .describe("AI backend to use: 'glm' (GLM-4.6V, default), 'kimi' (Kimi K2.5), or 'gemini' (Gemini 3 Flash Preview)"),
+        }),
+    }, searchTimestampTool);
+    // Tool 5: get_video_info - Get video metadata
+    server.registerTool('get_video_info', {
+        title: 'Get Video Info',
+        description: 'Get video metadata including duration, resolution, fps, codec, file size, and format. Supports local files (including file:// URIs) and remote http(s) URLs. No AI backend required.',
+        inputSchema: z.object({
+            videoPath: z
+                .string()
+                .describe('Path to the video file (local path, file:// URI, or http(s) URL)'),
+        }),
+    }, getVideoInfoTool);
+    // Tool 6: transcribe_video - Transcribe audio from a video
+    server.registerTool('transcribe_video', {
+        title: 'Transcribe Video',
+        description: 'Extract audio from a video and transcribe it using a dedicated speech-to-text provider (Deepgram, AssemblyAI, Groq/Whisper, or Gemini). Supports speaker diarization and translation to English.',
+        inputSchema: z.object({
+            videoPath: z
+                .string()
+                .describe('Path to the video file (local path, file:// URI, or http(s) URL)'),
+            provider: z
+                .enum(['deepgram', 'assemblyai', 'groq', 'gemini'])
+                .optional()
+                .describe("Audio provider to use: 'deepgram' (Nova-2, default), 'assemblyai' (Universal), 'groq' (Whisper-large-v3, free), or 'gemini'. Defaults to AUDIO_MCP_DEFAULT_PROVIDER env var or first available key."),
+            language: z
+                .string()
+                .optional()
+                .describe("BCP-47 language code, e.g. 'en', 'es', 'fr'. Auto-detected if omitted."),
+            diarize: z
+                .boolean()
+                .optional()
+                .describe('Enable speaker diarization (who said what). Supported by Deepgram and AssemblyAI only. Silently ignored for other providers.'),
+            translate: z
+                .boolean()
+                .optional()
+                .describe('Translate the transcript to English. Supported by Groq and Gemini only. Silently ignored for other providers.'),
+        }),
+    }, transcribeVideoTool);
+    // Connect to VS Code via stdio transport
+    const transport = new StdioServerTransport();
+    await server.connect(transport);
+    // Server is now running, listening for tool calls from Copilot
+}
+main().catch((error) => {
+    console.error('Fatal error starting video-mcp server:', error);
+    process.exit(1);
+});
+//# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":";AAEA,OAAO,EAAE,SAAS,EAAE,MAAM,yCAAyC,CAAA;AACnE,OAAO,EAAE,oBAAoB,EAAE,MAAM,2CAA2C,CAAA;AAChF,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAA;AAEvB,uBAAuB;AACvB,OAAO,EAAE,gBAAgB,EAAE,MAAM,yBAAyB,CAAA;AAC1D,OAAO,EAAE,kBAAkB,EAAE,MAAM,2BAA2B,CAAA;AAC9D,OAAO,EAAE,iBAAiB,EAAE,MAAM,0BAA0B,CAAA;AAC5D,OAAO,EAAE,mBAAmB,EAAE,MAAM,4BAA4B,CAAA;AAChE,OAAO,EAAE,gBAAgB,EAAE,MAAM,yBAAyB,CAAA;AAC1D,OAAO,EAAE,mBAAmB,EAAE,MAAM,4BAA4B,CAAA;AAChE,OAAO,EAAE,eAAe,EAAE,MAAM,mBAAmB,CAAA;AAEnD;;;GAGG;AAEH,KAAK,UAAU,IAAI;IACjB,0CAA0C;IAC1C,MAAM,MAAM,GAAG,IAAI,SAAS,CAC1B;QACE,IAAI,EAAE,WAAW;QACjB,OAAO,EAAE,OAAO;KACjB,EACD;QACE,YAAY,EAAE;YACZ,OAAO,EAAE,EAAE,EAAE,wCAAwC;SACtD;KACF,CACF,CAAA;IAED,8DAA8D;IAC9D,eAAe,CAAC,MAAM,CAAC,CAAA;IAEvB,oCAAoC;IAEpC,4DAA4D;IAC5D,MAAM,CAAC,YAAY,CACjB,eAAe,EACf;QACE,KAAK,EAAE,eAAe;QACtB,WAAW,EACT,mGAAmG;QACrG,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC;YACpB,SAAS,EAAE,CAAC;iBACT,MAAM,EAAE;iBACR,QAAQ,CAAC,4CAA4C,CAAC;YACzD,QAAQ,EAAE,CAAC;iBACR,MAAM,EAAE;iBACR,QAAQ,CAAC,yCAAyC,CAAC;YACtD,QAAQ,EAAE,CAAC;iBACR,IAAI,CAAC,CAAC,KAAK,EAAE,MAAM,EAAE,QAAQ,CAAC,CAAC;iBAC/B,QAAQ,EAAE;iBACV,QAAQ,CACP,wGAAwG,CACzG;SACJ,CAAC;KACH,EACD,gBAAgB,CACjB,CAAA;IAED,8DAA8D;IAC9D,MAAM,CAAC,YAAY,CACjB,iBAAiB,EACjB;QACE,KAAK,EAAE,iBAAiB;QACxB,WAAW,EACT,8MAA8M;QAChN,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC;YACpB,SAAS,EAAE,CAAC;iBACT,MAAM,EAAE;iBACR,QAAQ,CAAC,4CAA4C,CAAC;YACzD,QAAQ,EAAE,CAAC;iBACR,IAAI,CAAC,CAAC,KAAK,EAAE,MAAM,EAAE,QAAQ,CAAC,CAAC;iBAC/B,QAAQ,EAAE;iBACV,QAAQ,CACP,wGAAwG,CACzG;SACJ,CAAC;KACH,EACD,kBAAkB,CACnB,CAAA;IAED,qDAAqD;IACrD,MAAM,CAAC,YAAY,CACjB,gBAAgB,EAChB;QACE,KAAK,EAAE,gBAAgB;QACvB,WAAW,EACT,iKAAiK;QACnK,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC;YACpB,SAAS,EAAE,CAAC;iBACT,MAAM,EAAE;iBACR,QAAQ,CACP,kEAAkE,CACnE;YACH,IAAI,EAAE,CAAC;iBACJ,IAAI,CAAC,CAAC,MAAM,EAAE,UAAU,EAAE,YAAY,CAAC,CAAC;iBACxC,QAAQ,CACP,qHAAqH,CACtH;YACH,KAAK,EAAE,CAAC;iBACL,MAAM,EAAE;iBACR,GAAG,EAAE;iBACL,GAAG,CAAC,CAAC,CAAC;iBACN,GAAG,CAAC,GAAG,CAAC;iBACR,QAAQ,EAAE;iBACV,QAAQ,CAAC,wDAAwD,CAAC;YACrE,WAAW,EAAE,CAAC;iBACX,MAAM,EAAE;iBACR,GAAG,CAAC,GAAG,CAAC;iBACR,QAAQ,EAAE;iBACV,QAAQ,CACP,mEAAmE,CACpE;YACH,UAAU,EAAE,CAAC;iBACV,KAAK,CAAC,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC;iBACxB,QAAQ,EAAE;iBACV,QAAQ,CACP,iEAAiE,CAClE;SACJ,CAAC;KACH,EACD,iBAAiB,CAClB,CAAA;IAED,kEAAkE;IAClE,MAAM,CAAC,YAAY,CACjB,kBAAkB,EAClB;QACE,KAAK,EAAE,kBAAkB;QACzB,WAAW,EACT,0LAA0L;QAC5L,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC;YACpB,SAAS,EAAE,CAAC;iBACT,MAAM,EAAE;iBACR,QAAQ,CACP,kEAAkE,CACnE;YACH,KAAK,EAAE,CAAC;iBACL,MAAM,EAAE;iBACR,QAAQ,CACP,mEAAmE,CACpE;YACH,QAAQ,EAAE,CAAC;iBACR,IAAI,CAAC,CAAC,KAAK,EAAE,MAAM,EAAE,QAAQ,CAAC,CAAC;iBAC/B,QAAQ,EAAE;iBACV,QAAQ,CACP,wGAAwG,CACzG;SACJ,CAAC;KACH,EACD,mBAAmB,CACpB,CAAA;IAED,8CAA8C;IAC9C,MAAM,CAAC,YAAY,CACjB,gBAAgB,EAChB;QACE,KAAK,EAAE,gBAAgB;QACvB,WAAW,EACT,sLAAsL;QACxL,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC;YACpB,SAAS,EAAE,CAAC;iBACT,MAAM,EAAE;iBACR,QAAQ,CACP,kEAAkE,CACnE;SACJ,CAAC;KACH,EACD,gBAAgB,CACjB,CAAA;IAED,2DAA2D;IAC3D,MAAM,CAAC,YAAY,CACjB,kBAAkB,EAClB;QACE,KAAK,EAAE,kBAAkB;QACzB,WAAW,EACT,kMAAkM;QACpM,WAAW,EAAE,CAAC,CAAC,MAAM,CAAC;YACpB,SAAS,EAAE,CAAC;iBACT,MAAM,EAAE;iBACR,QAAQ,CACP,kEAAkE,CACnE;YACH,QAAQ,EAAE,CAAC;iBACR,IAAI,CAAC,CAAC,UAAU,EAAE,YAAY,EAAE,MAAM,EAAE,QAAQ,CAAC,CAAC;iBAClD,QAAQ,EAAE;iBACV,QAAQ,CACP,qMAAqM,CACtM;YACH,QAAQ,EAAE,CAAC;iBACR,MAAM,EAAE;iBACR,QAAQ,EAAE;iBACV,QAAQ,CACP,wEAAwE,CACzE;YACH,OAAO,EAAE,CAAC;iBACP,OAAO,EAAE;iBACT,QAAQ,EAAE;iBACV,QAAQ,CACP,8HAA8H,CAC/H;YACH,SAAS,EAAE,CAAC;iBACT,OAAO,EAAE;iBACT,QAAQ,EAAE;iBACV,QAAQ,CACP,+GAA+G,CAChH;SACJ,CAAC;KACH,EACD,mBAAmB,CACpB,CAAA;IAED,yCAAyC;IACzC,MAAM,SAAS,GAAG,IAAI,oBAAoB,EAAE,CAAA;IAC5C,MAAM,MAAM,CAAC,OAAO,CAAC,SAAS,CAAC,CAAA;IAE/B,+DAA+D;AACjE,CAAC;AAED,IAAI,EAAE,CAAC,KAAK,CAAC,CAAC,KAAK,EAAE,EAAE;IACrB,OAAO,CAAC,KAAK,CAAC,wCAAwC,EAAE,KAAK,CAAC,CAAA;IAC9D,OAAO,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;AACjB,CAAC,CAAC,CAAA"}

package/dist/services/audio/assemblyAiClient.d.ts

@@ -0,0 +1,28 @@
+/**
+ * AssemblyAI Audio Client
+ * Transcribes audio files using AssemblyAI's Universal (best) model
+ * Supports speaker diarization and other audio intelligence features
+ */
+export interface AssemblyAiTranscribeOptions {
+    language?: string;
+    diarize?: boolean;
+}
+export interface TranscriptSegment {
+    speaker?: string;
+    text: string;
+}
+export interface TranscriptResult {
+    text: string;
+    segments?: TranscriptSegment[];
+}
+export declare class AssemblyAiClient {
+    private client;
+    constructor(apiKey: string);
+    /**
+     * Transcribe an audio file using AssemblyAI
+     * @param audioPath - Local path to the audio file (.m4a, .mp3, etc.)
+     * @param options - Transcription options
+     */
+    transcribe(audioPath: string, options?: AssemblyAiTranscribeOptions): Promise<TranscriptResult>;
+}
+//# sourceMappingURL=assemblyAiClient.d.ts.map

package/dist/services/audio/assemblyAiClient.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"assemblyAiClient.d.ts","sourceRoot":"","sources":["../../../src/services/audio/assemblyAiClient.ts"],"names":[],"mappings":"AAEA;;;;GAIG;AAEH,MAAM,WAAW,2BAA2B;IAC1C,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,OAAO,CAAC,EAAE,OAAO,CAAA;CAClB;AAED,MAAM,WAAW,iBAAiB;IAChC,OAAO,CAAC,EAAE,MAAM,CAAA;IAChB,IAAI,EAAE,MAAM,CAAA;CACb;AAED,MAAM,WAAW,gBAAgB;IAC/B,IAAI,EAAE,MAAM,CAAA;IACZ,QAAQ,CAAC,EAAE,iBAAiB,EAAE,CAAA;CAC/B;AAED,qBAAa,gBAAgB;IAC3B,OAAO,CAAC,MAAM,CAAY;gBAEd,MAAM,EAAE,MAAM;IAI1B;;;;OAIG;IACG,UAAU,CACd,SAAS,EAAE,MAAM,EACjB,OAAO,GAAE,2BAAgC,GACxC,OAAO,CAAC,gBAAgB,CAAC;CAkC7B"}