Skip to content
Scalekit Docs
Talk to an Engineer Dashboard

Connect AI agents to the Readwise MCP server

Vendor MCP
Open markdown

The Readwise MCP connector routes your AI agent's tool calls to Readwise's own MCP server through Scalekit. Each user signs in to Readwise once, and Scalekit stores and refreshes their tokens, so your agent never handles credentials. It comes with 22 tools.

Tools
22
What they doRead · write · destructive
9 · 6 · 79 read6 write7 destructive
Users sign in with
OAuth app
Your own Readwise MCP server app

Setup

  1. Install the SDK

    Terminal window
    npm install @scalekit-sdk/node dotenv
  2. Set your credentials

    Add your Scalekit credentials to your .env file. Find values in app.scalekit.com > Developers > API Credentials.

    .env
    SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
    SCALEKIT_CLIENT_ID=<your-client-id>
    SCALEKIT_CLIENT_SECRET=<your-client-secret>
  3. Create the Readwise MCP connection

    In AgentKit > Connections, create a Readwise MCP connection and copy its redirect URI. The name you give it is the connection_name your code passes. See Configure connections.

  4. Register an OAuth app

    Readwise MCP server connections use your own OAuth app. Register one with Readwise MCP server and add the redirect URI you copied.

    Then enter the app's Client ID and Client Secret on the Readwise MCP connection.

  5. Authorize a user and make your first call

    quickstart.mts
    import { ScalekitClient } from '@scalekit-sdk/node'
    import 'dotenv/config'
    import { createInterface } from 'node:readline/promises'
    const scalekit = new ScalekitClient(
    process.env.SCALEKIT_ENVIRONMENT_URL,
    process.env.SCALEKIT_CLIENT_ID,
    process.env.SCALEKIT_CLIENT_SECRET,
    )
    const actions = scalekit.actions
    const connector = 'readwisemcp'
    const identifier = 'user_123'
    // Generate an authorization link for the user
    const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
    console.log('Authorize Readwise MCP:', link)
    const rl = createInterface({ input: process.stdin, output: process.stdout })
    await rl.question('Press Enter after authorizing...')
    rl.close()
    // Make your first call
    const result = await actions.executeTool({
    connector,
    identifier,
    toolName: 'readwisemcp_reader_list_documents',
    toolInput: {},
    })
    console.log(result)
    Terminal window
    npx tsx quickstart.mts

    Each user signs in once. See Authorize a user for the full flow and statuses.

Tools

Pass the exact name to execute_tool
Try in PlaygroundRequest a tool
  • readwisemcp_reader_get_document_detailsGet Reader document details, including Markdown contentRead-only

    Get Reader document details

    Get Reader document details, including Markdown content

    Inputs

    document_idstringrequired
    The ID of the document to get details for
  • readwisemcp_reader_get_document_highlightsGet highlights for a Reader documentRead-only

    Get Reader document highlights

    Get highlights for a Reader document

    Inputs

    document_idstringrequired
    The ID of the document to get highlights for
  • readwisemcp_reader_get_export_documents_statusCheck the status of a previously triggered Reader document export.Read-only

    Check Reader export status

    Check the status of a previously triggered Reader document export. Returns `status` ('processing' or 'completed'), progress counts, and a `download_url` when the export is complete.

    Inputs

    export_idstringrequired
    The export ID returned by `reader_export_documents`.
  • readwisemcp_reader_list_documentsList Reader documents.Read-only

    List Reader documents

    List Reader documents. Documents are returned from the most recently added to the least recently added. This tool can be used to find N newest documents. Tip: unseen documents have first_opened_at=null, seen documents have a non-null first_opened_at. You can mark documents as seen with reader_bulk_edit_document_metadata(documents=[{document_id: ..., seen: true}]). Tip: for videos (e.g. YouTube) and podcasts, word_count and reading_time are null; use listening_time (duration formatted like reading_time, e.g. '13 mins') instead.

    Inputs

    categorystring
    Filter by document categoryone of articleemailrsshighlightnotepdfepubtweetvideopodcastaudiobook
    idstring
    The document's unique ID. Using this parameter will return just one document, if found.
    limitinteger
    The maximum number of documents to returndefault 10
    locationstring
    Filter by document location. 'new' = inbox (where newly saved documents go), 'later' = saved for later, 'shortlist' = shortlisted, 'archive' = archived, 'feed' = RSS feed items (only use when user explicitly asks about feed/RSS). When user says 'inbox', use 'new'.one of newlatershortlistarchivefeed
    page_cursorstring
    A string returned by a previous request to get the next page of documents
    response_fieldsarray
    Optional fields to include in each document result. The 'id' field is always included. If not specified, all fields except the full-body 'content' and 'html_content' are included, since a single document body can be 100k+ characters. Request 'content'/'html_content' explicitly here (or use reader_get_document_details) only when you need the full text of the listed documents.
    seenboolean
    Filter by seen status. True = only documents that have been opened, False = only unopened documents.
    tagarray
    Filter by tag(s). Pass up to 5 tags to find documents having all the tags listed. Pass empty string in the list to find untagged documents.
    updated_afterstring
    Fetch only documents updated after this date (ISO 8601 datetime)
  • readwisemcp_reader_list_tagsList Reader tagsRead-only

    List Reader tags

    List Reader tags

    Inputs

    This tool takes no inputs.

  • readwisemcp_reader_search_documentsSearch Reader documents using hybrid searchRead-only

    Search Reader documents

    Search Reader documents using hybrid search

    Inputs

    querystringrequired
    Search in document content
    author_searchstring
    Search document author
    category_inarray
    Restrict search to specific categories
    document_idstring
    Restrict search to a specific document
    limitinteger
    The maximum number of documents to returndefault 20
    location_inarray
    Restrict search to specific locations. Values: 'new' (inbox), 'later', 'shortlist', 'archive'. When user says 'inbox', use ['new']. This tool does not search the feed, only locations in the user's library.Defaults to ["new", "later", "shortlist", "archive"].
    note_searchstring
    Search in document notes
    published_datestring
    Search by published date (ISO 8601 date)
    published_date_gtstring
    Search by published date greater than (ISO 8601 date)
    published_date_gtestring
    Search by published date greater than or equal to (ISO 8601 date)
    published_date_ltstring
    Search by published date less than (ISO 8601 date)
    published_date_ltestring
    Search by published date less than or equal to (ISO 8601 date)
    source_searchstring
    Search in document source
    summary_searchstring
    Search in document summary
    tags_inarray
    Restrict search to specific tags
    tags_searchstring
    Search in document tags
    title_searchstring
    Search in document title
    url_searchstring
    Search in document URL
  • readwisemcp_readwise_get_daily_reviewGet the user's daily review of highlights.Read-only

    Get Readwise daily review

    Get the user's daily review of highlights. Returns today's review with all highlights selected for review by the spaced repetition algorithm. Includes the review URL where the user can complete the review interactively.

    Inputs

    This tool takes no inputs.

  • readwisemcp_readwise_list_highlightsList Readwise highlights with optional filters.Read-only

    List Readwise highlights

    List Readwise highlights with optional filters. Highlights are returned from the most recently added to the least recently added. This tool can be used to find N newest highlights. Use `response_fields` to control which fields are returned — omit it for the default set (all base highlight fields, no book detail), or pass a subset (and optionally any book_* field) to minimize the response and include parent-book metadata.

    Inputs

    book_idinteger
    Filter by book ID
    highlighted_at_gtstring
    Filter highlights created after this datetime (ISO 8601)
    highlighted_at_ltstring
    Filter highlights created before this datetime (ISO 8601)
    pageinteger
    Page number for pagination
    page_sizeinteger
    Number of highlights per pagedefault 100
    response_fieldsarray
    Fields to include on each highlight result. Use this to reduce response size and save tokens. The 'id' field is always included. If not specified, every base highlight field is returned (including 'book_id') but no book-detail fields (book_title, book_author, ...) are fetched. Requesting any book-detail field enriches each highlight with its parent book's metadata in the same request.
    updated_gtstring
    Filter highlights updated after this datetime (ISO 8601)
    updated_ltstring
    Filter highlights updated before this datetime (ISO 8601)