RSSMonster 2.2.0

RSSMonster 2.2.0 focuses on faster article browsing, more reliable crawling, safer multi-user behavior, and substantially improved documentation.

Highlights

Added database-backed cursor pagination for supported article views, reducing the amount of data loaded up front and making large collections faster and more stable to browse.
Added detection for articles that arrive while a paginated view is open, with a refresh path that keeps the current list snapshot consistent.
Improved article, smart-folder, tag, category, overview, and Interest Island query performance, including bounded counts and fewer repeated lookups.
Added a Retry crawl action to feed settings so an individual feed can be crawled again through the normal production pipeline.
Added crawl-run heartbeats, stale-run recovery, and richer completion statistics to make interrupted and long-running crawls easier to detect and recover.

Fixed HTTP retry attempt counting.
Fixed category reordering.
Fixed concurrent article click-count updates.
Fixed RSS API limit handling.
Fixed a race during first-user administrator assignment.
Improved article-list loading, refresh, selection, read-state, and stale-response handling.
Reduced unnecessary article-feed watchers and improved client-side state synchronization.
Improved empty-reader, mobile navigation, pull-to-refresh, feed refresh, and settings error handling.

Hardened the MCP endpoint configuration and client error handling.
Added safer handling for external article and assistant source links.
Improved frontend URL validation and browser-storage protection.
Added a minimum cookie-availability check to authentication startup.
Tightened ownership and authorization checks across article and feed operations.

Added a new GitHub Pages documentation site using Just the Docs.
Added or expanded guides for installation, configuration, administration, crawling, APIs, bookmarks, tags, keyboard shortcuts, scoring, search, server jobs, and article embeddings.
Added in-depth documentation for Events, Interest Islands, Topics, and the semantic processing architecture.
Updated dependencies and project metadata for version 2.2.0.

This release includes database migrations for atomic first-user administrator claiming and crawl-run heartbeat tracking. Run the normal database migration step when upgrading if your deployment does not do so automatically.
No intentional breaking API changes are included.

Full comparison: v2.1.1...v2.2.0