paladini/mcp-me

Changed

• Medium generator — now uses a 4-tier fetch strategy to retrieve all published articles instead of only the last 10 from RSS: (1) unofficial JSON endpoint (?format=json) extracts the full post list with word counts, subtitles, and tags, with full pagination via payload.paging.path + payload.paging.next; (2) sitemap XML extraction for robust URL discovery; (3) HTML scraping parses __NEXT_DATA__, Apollo state, or raw href links as fallback; (4) RSS is used only as a last resort (capped at 10 by Medium). Removed the hard-coded .slice(0, 10) limit and updated request headers to browser-like values to reduce 403 blocks.
• Generator docs and templates — documentation now covers the native --blogger-backup generator in English, including what it is for, how to obtain a Blogger XML export, and how to filter multi-author backups.

Added

• Blogger XML backup generator — added a native --blogger-backup generator that reads a local Blogger XML export, optionally filters posts by author name/email using path::author1,author2,... syntax, imports matching posts into projects.yaml with category: article, preserves labels as tags, and adds archive summary FAQ entries.