This MCP server integrates with the Pocket API, allowing AI assistants to interact with users' saved articles. Developed by kazuph, it enables fetching unread articles and marking them as read through natural language commands. The TypeScript implementation uses the Model Context Protocol SDK and Zod for robust type checking. It's designed for Pocket users who want to incorporate AI assistance into their reading workflow, enabling tasks like summarizing saved articles, prioritizing reading lists, or analyzing reading habits through conversational interfaces.
Fetch your saved articles from Pocket API. Returns title, URL, and excerpt for each article.
Mark a specific article as read (archived) in your Pocket account using its item ID.
No reviews yet. Be the first to review!
Sign in to join the conversation
Highlight AIStart the server with node to access it via any client or IDE.
node path/to/downloaded/file.mjs