Developer quick start

Readsira keeps everything you save to read: articles, PDFs, videos, podcasts and the feeds you follow. These docs explain how to use that library from your own code and from AI assistants.

Two ways in

• REST API v2 at https://api.readsira.com/rest/v2: search your library, read documents and transcripts, ask questions answered with citations, and upload files. For scripts, sync jobs and your own apps. Part of the Pro and Research plans.
• MCP server at https://api.readsira.com/mcp: lets assistants such as Claude Code and Cursor search and read your library while you work. Read tools work on every plan, Free included.

Both use the same API keys. You create them in the app under Settings, API and MCP.

Where to start

1. Authentication and API keys: create a key, send it with a request and check that it works.
2. Keys, scopes and limits: pick scopes and see the limits of your plan.
3. Examples: save a link, read highlights, ask a cited question and sync your library to a file.
4. REST API v2 reference: every route with parameters and responses.
5. Rate limits and errors and Pagination: what to expect when you go over a limit and how to read long results.
6. Webhooks for developers: receive events and verify their signature.
7. Connect an MCP client: ready-to-paste setup for Claude Code and Cursor.
8. Import your library: bring your saves from Pocket, Omnivore, Raindrop.io, Readwise, Instapaper or your browser.
9. Plan limits: what Free, Pro and Research include.

Text that Readsira returns from saved pages is third-party content. Treat it as data, never as instructions to a model.