mirror of
https://github.com/RooVetGit/Roo-Code.git
synced 2026-10-07 02:58:15 +00:00
97 lines
3.2 KiB
Markdown
97 lines
3.2 KiB
Markdown
# Marketplace Data
|
|
|
|
This directory contains marketplace data files for Roo Code's marketplace system.
|
|
|
|
## Structure
|
|
|
|
- `mcps.yaml` - MCP (Model Context Protocol) servers available in the marketplace
|
|
- `modes.yaml` - Custom modes available in the marketplace (to be added)
|
|
|
|
## MCP Servers
|
|
|
|
The `mcps.yaml` file contains a list of MCP servers that can be installed through the Roo Code marketplace. Each MCP server entry includes:
|
|
|
|
- `id` - Unique identifier for the MCP server
|
|
- `name` - Display name
|
|
- `description` - Detailed description of functionality
|
|
- `author` - Author name
|
|
- `authorUrl` - Author's website or GitHub profile
|
|
- `url` - Repository or documentation URL
|
|
- `tags` - Array of tags for categorization
|
|
- `prerequisites` - Array of requirements needed before installation
|
|
- `content` - Installation methods (can be a single string or array of methods)
|
|
|
|
### Installation Methods
|
|
|
|
Each installation method can include:
|
|
|
|
- `name` - Method name (e.g., "uvx", "pip install")
|
|
- `content` - JSON configuration to be added to MCP settings
|
|
- `parameters` - Optional parameters that users can customize
|
|
- `prerequisites` - Method-specific prerequisites
|
|
|
|
### Parameters
|
|
|
|
Parameters allow users to customize the installation:
|
|
|
|
- `name` - Display name for the parameter
|
|
- `key` - Key used in the configuration template
|
|
- `placeholder` - Default value or example
|
|
- `optional` - Whether the parameter is optional (default: false)
|
|
|
|
## VoiceVox MCP Server
|
|
|
|
The VoiceVox MCP Server provides Japanese text-to-speech capabilities using the VoiceVox engine. It includes:
|
|
|
|
### Features
|
|
|
|
- Text-to-Speech conversion for Japanese text
|
|
- Multiple voice characters with different personalities
|
|
- Customizable speech speed and voice selection
|
|
- Automatic audio playback after generation
|
|
- WAV format audio file management
|
|
|
|
### Available Tools
|
|
|
|
1. `get_voices` - Retrieve list of available voices from VoiceVox
|
|
2. `text_to_speech` - Convert text to speech with customizable settings
|
|
|
|
### Use Cases
|
|
|
|
- Educational content creation with Japanese narration
|
|
- Accessibility improvements for Japanese content
|
|
- Podcast and video production
|
|
- Language learning applications
|
|
- Chatbot voice functionality
|
|
|
|
### Installation Options
|
|
|
|
1. **uvx (Recommended)** - Uses uvx to run the server directly
|
|
2. **pip install** - Install via pip and run with Python
|
|
3. **Custom Configuration** - Full customization with all available parameters
|
|
|
|
### Prerequisites
|
|
|
|
- VoiceVox Engine running (default: http://localhost:50021)
|
|
- Python 3.10+
|
|
- For uvx method: `pip install uvx`
|
|
- For pip method: `pip install mcp-server-voicevox`
|
|
|
|
## Usage
|
|
|
|
This marketplace data is used by the Roo Code extension to populate the marketplace UI. Users can browse, search, and install MCP servers directly from the extension.
|
|
|
|
## Contributing
|
|
|
|
To add a new MCP server to the marketplace:
|
|
|
|
1. Add an entry to `mcps.yaml` following the schema
|
|
2. Ensure all required fields are provided
|
|
3. Test the installation methods
|
|
4. Submit a pull request
|
|
|
|
## References
|
|
|
|
- [VoiceVox MCP Server Repository](https://github.com/Sunwood-ai-labs/mcp-voicevox)
|
|
- [VoiceVox MCP Server PyPI Package](https://pypi.org/project/mcp-server-voicevox/)
|
|
- [Model Context Protocol Documentation](https://docs.roocode.com/advanced-usage/mcp)
|