codevideo-mcp
1
codevideo-mcp is a Model Context Protocol (MCP) server designed to create educational content about software using natural language. It supports generating video lessons, blog posts, and verifying JSON actions.
codevideo-mcp
The MCP server for CodeVideo.
Create software educational content using natural language.
Examples:
Creation:
- "Build a video lesson on how to use the
useState
hook in React" - "Can you create a video lesson compare and contrasting how to program concurrently in C#, Go, and Rust?"
- Can you give me a markdown blog post on how to use the
useState
hook in React?"
Conversion:
- "Please convert these CodeVideo JSON actions to a video"
- "Convert these CodeVideo JSON actions to a blog post"
- "Convert these CodeVideo JSON actions to HTML"
Validation:
- "Are these CodeVideo JSON actions valid?"
Usage with Claude Desktop
Simply drop the following into your claude_desktop_config.json
file:
{
// ... other MCP servers
"codevideo-mcp": {
"command": "npx",
"args": [
"-y",
"@fullstackcraftllc/codevideo-mcp@latest"
],
"env": {
"ELEVENLABS_API_KEY": "your-elevenlabs-api-key",
"ELEVENLABS_VOICE_ID": "your-elevenlabs-voice-id"
}
},
// ... other MCP servers
}
Be sure to restart Claude after modifying this config!
After restarting Claude Desktop, if installation was successful, you should see codevideo-mcp
in the list of installed MCPs:
Local Development
Clone this repository:
git clone https://github.com/codevideo/codevideo-mcp.git
Then, install the dependencies and build the project:
cd codevideo-mcp
npm install
npm run build
Then, use the following JSON to run the local version on your Claude Desktop:
// ... other MCP servers
"codevideo-mcp": {
"command": "npx",
"args": [
"tsx",
"/path/to/your/clone/of/codevideo-mcp/src/index.ts"
],
"env": {
"PATH_TO_CODEVIDEO_CLI": "/path/to/your/codevideo-cli",
}
}
// ... other MCP servers
Known Issues
- Currently, creating entire courses seems to be a bit flakey and perhaps too large for the context window for Claude. We're working on refining the full course flow. Single lesson generation seems to work well.