Help us improve
Share bugs, ideas, or general feedback.
From media-mcp
Get started with media-metadata-mcp -- what it is, how to set it up, and how to use it
npx claudepluginhub cameronsjo/media-mcp --plugin media-mcpHow this skill is triggered — by the user, by Claude, or both
Slash command
/media-mcp:onboardThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Guide the user through getting started with **media-metadata-mcp**.
Finalizes documentation and project metadata for MCP servers after implementation is complete. Audits surface area, updates README and agent protocol files.
Finalizes documentation and project metadata for an MCP server after implementation is complete. Audits surface area, updates README and agent protocol files.
Integrates NotebookLM via MCP CLI for external RAG, research synthesis, creating/managing notebooks/sources, and generating audio, cinematic video, slides, infographics, mind maps.
Share bugs, ideas, or general feedback.
Guide the user through getting started with media-metadata-mcp.
An MCP server that enriches book, movie, and TV show metadata for Obsidian vaults. It queries Open Library, Google Books, Goodreads, and TMDB, merges and normalizes the data, and returns Obsidian-ready YAML frontmatter. Supports both stdio and Streamable HTTP transports.
Check that the user has the following installed/configured:
node --version)Walk the user through initial setup:
Install dependencies:
npm install
Build the TypeScript:
npm run build
Configure the MCP server in Claude Code or Claude Desktop. Add to your MCP config:
{
"mcpServers": {
"media-metadata": {
"command": "node",
"args": ["/Users/cameron/Projects/media-mcp/dist/index.js"],
"env": {
"TMDB_API_KEY": "your-tmdb-api-key"
}
}
}
}
Book lookups work without any API keys (Open Library + Goodreads scraping). Movie/TV lookups require TMDB_API_KEY.
Guide the user through their first interaction with the product:
Start the server in HTTP mode for quick testing:
TMDB_API_KEY=your-key npm run dev:http
The server starts at http://127.0.0.1:3000/mcp.
Or connect via Claude Code with the MCP config above and ask:
"Look up the book Project Hail Mary by Andy Weir"
Claude will call lookup_book and return metadata including title, author, ISBN, genres, page count, ratings, and cover URL.
Follow up with:
"Generate Obsidian frontmatter for that"
Claude will call generate_frontmatter and return ready-to-paste YAML.
Point the user to the most important files for understanding the project:
src/index.ts -- MCP server entry point, transport selectionsrc/tools/ -- Tool implementations: lookup-book.ts, lookup-movie.ts, lookup-tv.ts, generate-frontmatter.ts, batch-lookup.tssrc/sources/ -- Data source clients: open-library.ts, google-books.ts, goodreads.ts, tmdb.tssrc/utils/config.ts -- Environment variable configurationsrc/cache/sqlite-cache.ts -- SQLite caching layerpackage.json -- Scripts, dependencies, engine requirementsDockerfile -- Container build for HTTP transport deploymentnpm run devnpm run dev:httpnpm run buildnpm testnpm run lintnpm run typecheckdocker build -t media-metadata-mcp . && docker run -e TMDB_API_KEY=your-key -p 3000:3000 media-metadata-mcp