Update SKILL.md with new CLI options and MCP documentation

SKILL.md

@@ -96,6 +96,9 @@ python3 videogen --model-list --high-vram    # >30GB VRAM
 python3 videogen --model-list --huge-vram    # >55GB VRAM
 python3 videogen --model-list --nsfw-friendly
 
+# Batch output (for scripts)
+python3 videogen --model-list --model-list-batch
+
 # Disable a model from auto selection
 python3 videogen --disable-model <ID_or_name>
 
@@ -132,6 +135,16 @@ python3 videogen --model wan_14b_t2v --prompt "..." --vram_limit 16
 python3 videogen --model wan_14b_t2v --prompt "..." --low_ram_mode
 ```
 
+#### Output Options
+
+```bash
+# Specify output directory for batch processing
+python3 videogen --model wan_14b_t2v --prompt "..." --output-dir /path/to/output
+
+# Auto-confirm prompts (useful for scripts)
+python3 videogen --model wan_14b_t2v --prompt "..." --yes
+```
+
 ### Auto Mode
 
 Auto mode is the easiest way for AI agents to generate content:
@@ -329,6 +342,11 @@ python3 videogen --video input.mp4 --transcribe --whisper-model large
 
 # Specify source language
 python3 videogen --video input.mp4 --transcribe --source-lang en
+
+# Audio chunking strategies for long videos
+python3 videogen --video input.mp4 --transcribe --audio-chunk overlap     # 60s chunks with 2s overlap
+python3 videogen --video input.mp4 --transcribe --audio-chunk word-boundary  # Split at word boundaries
+python3 videogen --video input.mp4 --transcribe --audio-chunk vad           # Skip silence with VAD
 ```
 
 ### Create Subtitles
@@ -630,6 +648,52 @@ print(f"Model: {data['model']}")
 
 ---
 
+## MCP Server Integration
+
+VideoGen provides an MCP (Model Context Protocol) server for programmatic access:
+
+```bash
+# Start the MCP server
+python3 videogen_mcp_server.py
+```
+
+### MCP Tools Available
+
+The MCP server exposes the following tools:
+
+| Tool | Description |
+|------|-------------|
+| `videogen_list_models` | List available models (supports filter and batch parameters) |
+| `videogen_show_model` | Show model details by ID or name |
+| `videogen_generate_video` | Generate video from text (T2V) |
+| `videogen_generate_image` | Generate image from text (T2I) |
+| `videogen_animate_image` | Animate image (I2V) |
+| `videogen_transform_image` | Transform image (I2I) |
+| `videogen_generate_with_audio` | Generate video with audio |
+| `videogen_transcribe_video` | Transcribe video audio |
+| `videogen_create_subtitles` | Create subtitles |
+| `videogen_dub_video` | Dub/translate video |
+| `videogen_search_models` | Search HuggingFace for models |
+| `videogen_add_model` | Add custom model |
+| `videogen_update_models` | Update model database |
+
+### MCP Parameters
+
+New MCP parameters for the tools:
+
+```python
+# List models with batch output
+{"filter": "t2v", "batch": true}
+
+# Generate with output directory
+{"model": "wan_14b_t2v", "prompt": "...", "output_dir": "/path/to/output", "yes": true}
+
+# Transcribe with audio chunking
+{"video": "input.mp4", "audio_chunk": "overlap"}  # or "word-boundary" or "vad"
+```
+
+---
+
 ## Best Practices for AI Agents
 
 1. **Update models first** - Run `--update-models` before first use