Readwise Reader
Save, list, organize, update, and delete documents in a Readwise Reader library.
Verdict
Common use cases
- Archive competitor blog posts for analysis
- Tag research articles by project phase
- Save meeting prep links in bulk
- Pull reading lists into sprint planning
- Delete outdated documents from shared library
Integration
- Vendor
- Readwise Reader
- Category
- other
- Auth
- API_KEY
- Tools
- 6
- Composio slug
readwise_reader
Tools
- Bulk Update Documents
Partially update up to 50 Reader documents in one request and return a success or error result for every document.
- Delete Documentdestructive
Permanently delete one Reader document by its ID.
- List Documents
List, filter, incrementally sync, or fetch a specific Reader document. Returns one page and a cursor for the next page.
- List Tags
List tags in the connected Reader library, returning one page and a cursor for the next page.
- Save Document
Save a URL or supplied HTML as a document in the connected Reader library, with optional metadata, location, and tags. Returns the Reader document ID needed by update and delete tools.
- Update Document
Partially update supported fields on one Reader document by ID; omitted fields remain unchanged.
Setup
Setup guide
- 11. Open your Switchy workspace and navigate to Settings > Integrations. 2. Find Readwise Reader in the MCP catalog and click Connect. 3. Log into your Readwise account at readwise.io/access_token to generate an API key. 4. Copy the token and paste it into the Switchy connection dialog, then click Authorize. 5. Switchy confirms the connection and shows which tools are available. 6. Open any Project and type '@Readwise Reader list my recent documents' to verify the integration works. 7. The MCP returns a paginated list of your saved articles, PDFs, and feeds. 8. Use the returned document IDs in update or delete commands to manage your library from chat.
What teammates see: by default, memories from Readwise Reader are scoped to the Project (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Projects have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
Save Research Links
@Readwise Reader save these URLs to my library and tag them 'Q1-research': [paste URLs]. Return the document IDs so I can update them later.
List Recent Articles
@Readwise Reader list the 20 most recent documents in my library, filtered by the tag 'product-ideas'. Include titles and URLs.
Bulk Tag Documents
@Readwise Reader add the tag 'sprint-12' to these document IDs: [paste IDs]. Confirm each update succeeded.
Archive Completed Reads
@Readwise Reader update these document IDs to archived status: [paste IDs]. Let me know if any fail.
Delete Outdated Content
@Readwise Reader delete these document IDs from my library: [paste IDs]. Confirm each deletion.
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@readwise-reader list all documents tagged 'research' from the last 30 days and show me their titles and reading progress
Found 14 documents tagged 'research' saved in the past month. Top items: 'The State of LLM Agents' (68% read, saved Jan 15), 'Designing for AI Collaboration' (unread, saved Jan 10), 'Context Window Strategies' (42% read, saved Jan 8). The list includes 6 articles, 5 PDFs, and 3 Twitter threads. Most are unread or partially completed. Use the document IDs if you want to update tags or archive any of these.
This example uses the List Documents tool with tag and date filters. Readwise Reader's API returns paginated results — if your library has hundreds of tagged items, you'll get a cursor to fetch the next page. The MCP requires an API key with read access to your library.
@readwise-reader save this article https://example.com/post/ai-workflows and tag it 'switchy' and 'mcp-integration'
Saved 'AI Workflows for Small Teams' to your Reader library with tags 'switchy' and 'mcp-integration'. Document ID: doc_abc123xyz. The article is now queued for processing — Readwise will extract the full text and any highlights you add will sync back to your account. You can update tags or archive this document later using the ID.
This example uses the Save Document tool to add a URL to your library with custom tags. The MCP returns a document ID immediately, but Readwise processes the content asynchronously — full text and metadata may take a few seconds to appear in your library. Requires write access to your Reader account.
@readwise-reader find all unread documents tagged 'urgent', summarize their topics, then bulk update them to add a 'reviewed-jan-2025' tag
Found 8 unread documents tagged 'urgent'. Topics span API rate limiting (3 docs), team onboarding workflows (2 docs), security compliance updates (2 docs), and one feature request template. Applied 'reviewed-jan-2025' tag to all 8 documents via bulk update. All operations succeeded — no errors returned. These documents remain unread in your library but are now marked for your January review cycle.
This example chains List Documents (with tag and read-status filters) with Bulk Update Documents to modify up to 50 items at once. The AI synthesizes the list before acting, giving you a summary to confirm intent. Bulk updates are atomic per document — if one fails (e.g., invalid ID), others still succeed. Requires write access.
Use-case deep-dives
When Reader MCP beats manual bookmarking for content curation
A 2-person editorial team curating a weekly industry newsletter saves 30-50 links per week from Slack, email, and Twitter. The Reader MCP's Save Document tool lets them push URLs directly from their AI workspace with tags and notes attached, skipping the browser-extension dance. List Documents with tag filters then pulls the week's queue into a single prompt for summarization or comparison. The trade-off: if your team already lives in Notion or Airtable for content ops, the MCP adds a sync step you don't need. But if you're already paying for Readwise and your curation happens in chat or docs, this MCP closes the loop without a third tool. Worth it when your archive hits 500+ items and search becomes the bottleneck.
Using Reader MCP to centralize support article research
A 6-person support team at a B2B SaaS company saves competitor docs, help articles, and user-submitted links into Reader as reference material. The Bulk Update Documents tool lets them retag 20-30 articles at once when a product launch shifts their taxonomy, and List Tags keeps the workspace's AI assistant aware of the current category structure. The MCP shines when your team references the same 200-400 articles repeatedly and needs fast keyword search across PDFs and web snapshots. It falls short if your support stack already indexes this content in Zendesk or Guru—then you're duplicating effort. The buying call: if Reader is your team's single source of truth for external references, the MCP makes that library queryable from your AI workspace without tab-switching.
When Reader MCP handles quarterly competitor deep-dives
A 4-person product team tracks 8-10 competitors by saving their changelog pages, blog posts, and feature announcements into Reader every quarter. The Save Document tool accepts raw HTML, so they can archive paywalled or login-gated pages that disappear between reviews. List Documents with date filters then surfaces everything new since the last roadmap session, and the AI workspace drafts a comparison table without re-reading 40 tabs. The limitation: if your competitive intel lives in a dedicated tool like Crayon or Klue, the MCP is redundant. But if you're bootstrapped or your research is ad-hoc, Reader's tagging and the MCP's bulk operations beat a folder of bookmarks. The threshold is around 100 saved competitor artifacts—below that, a spreadsheet is faster.
Frequently asked
What does the Readwise Reader MCP do in Switchy?
It lets your AI agents save URLs or HTML to your Readwise Reader library, fetch documents with filters, update metadata like tags and reading progress, and delete items. You can bulk-update up to 50 documents at once. This is useful for automating research workflows—agents can clip articles, tag them by topic, or sync reading lists without you opening Reader.
Do I need a Readwise subscription to use this MCP?
Yes. You need an active Readwise account and an API key from your Reader settings. The MCP authenticates with that key, so it inherits your subscription's document limits and features. If you're on a free trial, the MCP works but you're still bound by trial restrictions on document count and retention.
Can the MCP highlight text or add notes to documents?
No. The MCP updates document-level metadata—tags, reading progress, archive status—but it doesn't create or edit highlights or notes within a document. If you need to annotate passages, you'll still do that in the Reader app or via Readwise's separate highlights API, which this MCP doesn't expose.
Why use this instead of manually saving links in Reader?
Agents can save dozens of URLs in one go, apply consistent tagging rules, or sync documents from other tools without you copy-pasting. The List Documents tool also lets you filter by date or tag, so an agent can pull your unread articles and summarise them in a daily brief—something you'd script manually otherwise.
Who on my team should connect the Reader account?
Whoever owns the Readwise subscription and wants agents to manage their library. Reader doesn't have team accounts, so each person needs their own API key. If you share a workspace, each teammate connects their own Reader; agents can't cross-pollinate libraries unless you manually export and re-import documents.