# Omnivore (Read-Later) MCP > Omnivore (Read-Later) connects your saved articles and links to any AI client. Use this MCP to manage your personal reading list directly through conversation, letting your agent search, summarize, and categorize knowledge from all your curated sources. ## Overview - **Category:** productivity - **Price:** Free - **Tags:** read-it-later, content-curation, bookmarking, article-retrieval, knowledge-base, search-indexing ## Description You keep hundreds of interesting web links in read-it-later apps, but finding that one specific piece of information later is a pain. This connector turns your entire Omnivore library into an active source for your AI agent. Your agent can find content based on labels or folders; it doesn't just show you a list of titles, it retrieves the full text and author details so you can analyze it immediately. You can also instantly save new articles from any website without leaving your chat window. If you use Vinkius to connect this MCP, you gain an immediate knowledge base that lets your AI agent act on everything you've ever read. ## Tools ### get_article Fetches the full text and details for one specific article you saved. ### get_me Checks your current Omnivore account details and confirms your connection status. ### save_url Adds a new web link to your library instantly, without opening the Omnivore website. ### search_articles Filters and finds articles across your entire collection using labels or reading status filters. ## Prompt Examples **Prompt:** ``` Search my Omnivore library for unread articles about 'Machine Learning'. ``` **Response:** ``` I found 3 unread articles about Machine Learning. The most recent is 'Understanding Transformers' (slug: understanding-transformers). Would you like me to fetch the full content of any of these? ``` **Prompt:** ``` Fetch the full content of the article with slug 'mcp-guide' for username 'alex_dev'. ``` **Response:** ``` I've retrieved the article 'The Ultimate MCP Guide'. It covers the Model Context Protocol architecture and implementation details. Would you like a summary of the key points? ``` **Prompt:** ``` Save the URL 'https://blog.omnivore.app/p/getting-started' to my library. ``` **Response:** ``` Successfully saved the URL to your Omnivore library. It is now available in your inbox for later reading. ``` ## Capabilities ### undefined undefined ### undefined undefined ### undefined undefined ### undefined undefined ## Use Cases ### Reviewing old research notes A researcher needs to write a literature review on AI ethics. Instead of manually searching files, they ask their agent to use `search_articles` for 'AI' AND label:ethics AND is:unread. The agent finds the top five most relevant sources and pulls the full content using `get_article` so they can start drafting immediately. ### Saving inspiration during a meeting A marketer hears about a new trend and wants to save an article link. They use their agent to call `save_url`, passing the URL in real time, ensuring the source material is captured for later brainstorming without taking notes. ### Checking content gaps A content creator has a project scope and wants to verify if they've saved enough examples. They ask their agent to search using `search_articles` filtered by 'topic:UX design' and check the count, ensuring they have sufficient source material. ### Getting quick context on an old article A user remembers reading a great piece from two months ago but can't recall where. They ask their agent to search for keywords, and the agent uses `search_articles` followed by `get_article` to pull up the full text, solving the mystery instantly. ## Benefits - Instead of opening multiple tabs to search your bookmarks, the `search_articles` tool lets your agent filter your whole library using labels or status. You just ask for what you need. - Need deep context? Use `get_article` to pull the complete text and metadata from a saved piece, allowing your AI client to summarize key points directly in the chat. - When you stumble upon an interesting article while working, use `save_url`. You instantly add it to your library without ever having to switch applications. It's built right into your conversation flow. - You can confirm everything is set up correctly using `get_me` to verify your account details and make sure the connection works before you start processing data. - It turns a passive collection of links into an active knowledge base, letting you retrieve source material on demand when writing or researching. ## How It Works The bottom line is that you treat your entire reading list like an indexed database you talk to naturally. 1. Subscribe to this MCP on Vinkius and enter your Omnivore API Key. 2. Connect your preferred AI client (Claude, Cursor, etc.) to the active catalog link. 3. Direct your agent: ask it to search for a topic or save a new URL from your conversation. ## Frequently Asked Questions **How do I use Omnivore (Read-Later) MCP to search my articles?** You tell your agent to find what you need. It uses `search_articles` and accepts filters like labels or read status, letting you narrow down thousands of links to just the relevant ones. **Can I save new links using Omnivore (Read-Later) MCP?** Yes. You can use `save_url` right in your chat conversation to instantly add a link to your library without needing to visit the Omnivore site. **What does get_article do with Omnivore (Read-Later) MCP?** The `get_article` tool retrieves the complete, full text of an article. This is crucial because it lets your agent summarize or analyze content that you can't just skim from a title. **Do I need to use Omnivore (Read-Later) MCP with all my clients?** No, you only connect the MCP to the AI client you prefer. Vinkius lets your agent work whether it's connected from Cursor or Claude. **Is there a tool to check if my connection is working with Omnivore (Read-Later) MCP?** Yes, use `get_me`. This simple tool checks your account details and confirms that the API key you provided is active and connected.